| Bico_Fino | I use Zabbix | 00:01 |
|---|---|---|
| Bico_Fino | \o | 00:01 |
| Bico_Fino | Since 1.0.3. :D | 00:01 |
| gunix | we use bmc patrol at work | 00:01 |
| Bico_Fino | I saw bmc once. | 00:01 |
| gunix | ever heard of it? it's so bad, you are better off manually loging on servers | 00:01 |
| Bico_Fino | Not pretty. | 00:01 |
| gunix | yes, beauty is not one of it's strong points | 00:02 |
| cloudnull | mhayden: started https://github.com/openstack/monitorstack -- which i worked on a bit. | 00:02 |
| gunix | that's because it doesn't have any strong points | 00:02 |
| cloudnull | the aim was to be a check system that was backend agnostic | 00:02 |
| cloudnull | and my hope is to convert all of the rpc-maas checks to monitorstack plugins | 00:03 |
| cloudnull | but ive not got it done yet. | 00:03 |
| Bico_Fino | I think we should have an playbook with all endpoints and checks and sent to some monitoring stack? | 00:04 |
| cloudnull | ++ | 00:04 |
| gunix | lol contact info for monitorstack is this channel | 00:04 |
| cloudnull | that'd be lovely | 00:04 |
| cloudnull | gunix: it was something we started here. | 00:04 |
| Bico_Fino | I mean | 00:04 |
| Bico_Fino | openstack endpoint list - could be a start | 00:05 |
| gunix | i should go to sleep. it's already 2 am h ere | 00:05 |
| cloudnull | sleep is for the weak | 00:06 |
| cloudnull | MOAR CLOUD! | 00:06 |
| Bico_Fino | 10 pm here | 00:06 |
| Bico_Fino | :P | 00:06 |
| * cloudnull im out. | 00:06 | |
| cloudnull | bbl | 00:06 |
| cloudnull | going to feed dogs and !cloud for a while | 00:07 |
| gunix | dog person? | 00:07 |
| gunix | i have cats | 00:07 |
| gunix | but i live in an apartment with 3 rooms in the middle of a big city | 00:08 |
| Bico_Fino | setup-openstack.yml finished well | 00:09 |
| Bico_Fino | Going try to setup OVS tomorrow | 00:10 |
| cloudnull | yup, i've two dogs. | 00:11 |
| cloudnull | camilo https://twitter.com/cloudnull/status/915559522336428032 | 00:11 |
| cloudnull | felipe https://twitter.com/cloudnull/status/904104707928547328 | 00:12 |
| *** yifei has joined #openstack-ansible | 00:18 | |
| *** oanson has quit IRC | 00:22 | |
| *** oanson has joined #openstack-ansible | 00:24 | |
| *** chas has quit IRC | 00:25 | |
| *** chas has joined #openstack-ansible | 00:25 | |
| *** pbandark has quit IRC | 00:30 | |
| *** chas has quit IRC | 00:30 | |
| *** snowman48391 has quit IRC | 00:31 | |
| *** dxiri has quit IRC | 00:37 | |
| *** dxiri has joined #openstack-ansible | 00:37 | |
| *** lbragstad has joined #openstack-ansible | 00:39 | |
| *** vnogin has joined #openstack-ansible | 00:44 | |
| *** lbragstad has quit IRC | 00:47 | |
| *** snowman4839 has joined #openstack-ansible | 00:47 | |
| *** xingchao has joined #openstack-ansible | 00:47 | |
| *** a0x3e8 has quit IRC | 00:49 | |
| *** jiaopengju has joined #openstack-ansible | 00:51 | |
| *** vnogin has quit IRC | 00:51 | |
| *** woodard has quit IRC | 00:52 | |
| *** woodard has joined #openstack-ansible | 00:53 | |
| *** xingchao has quit IRC | 01:05 | |
| *** vnogin has joined #openstack-ansible | 01:06 | |
| *** gouthamr has quit IRC | 01:08 | |
| *** vnogin has quit IRC | 01:10 | |
| *** vnogin has joined #openstack-ansible | 01:11 | |
| *** vnogin has quit IRC | 01:12 | |
| *** xingchao has joined #openstack-ansible | 01:13 | |
| *** vnogin has joined #openstack-ansible | 01:16 | |
| *** hw_wutianwei has joined #openstack-ansible | 01:17 | |
| *** vnogin has quit IRC | 01:21 | |
| *** vnogin has joined #openstack-ansible | 01:26 | |
| *** vnogin has quit IRC | 01:30 | |
| *** vnogin has joined #openstack-ansible | 01:31 | |
| *** xingchao has quit IRC | 01:35 | |
| *** vnogin has quit IRC | 01:47 | |
| *** oanson has quit IRC | 01:52 | |
| *** oanson has joined #openstack-ansible | 01:52 | |
| *** lbragstad has joined #openstack-ansible | 02:08 | |
| *** galstrom_zzz is now known as galstrom | 02:12 | |
| *** vnogin has joined #openstack-ansible | 02:12 | |
| *** Miouge has quit IRC | 02:26 | |
| *** Jeffrey4l has quit IRC | 02:26 | |
| *** chas has joined #openstack-ansible | 02:27 | |
| *** Jeffrey4l has joined #openstack-ansible | 02:29 | |
| *** chas has quit IRC | 02:31 | |
| *** lbragstad has quit IRC | 02:34 | |
| *** sxc731 has joined #openstack-ansible | 02:42 | |
| *** galstrom is now known as galstrom_zzz | 02:44 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_glance master: Add external LB management handler hook interface https://review.openstack.org/504619 | 02:55 |
| *** thorst has joined #openstack-ansible | 02:55 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_barbican master: Update paste, policy and rootwrap configurations 2017-11-02 https://review.openstack.org/517519 | 02:57 |
| *** thorst has quit IRC | 02:58 | |
| *** thorst has joined #openstack-ansible | 02:58 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Update all SHAs for 16.0.1 https://review.openstack.org/517520 | 03:02 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Update all SHAs for master https://review.openstack.org/517520 | 03:03 |
| *** armaan has quit IRC | 03:09 | |
| *** armaan has joined #openstack-ansible | 03:09 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible stable/pike: Update swift role SHA https://review.openstack.org/517522 | 03:11 |
| *** thorst has quit IRC | 03:13 | |
| openstackgerrit | Merged openstack/openstack-ansible-repo_server master: Ensure the /var/log/apt-cacher-ng directory created https://review.openstack.org/517107 | 03:18 |
| *** yifei has quit IRC | 03:19 | |
| *** mwynne has quit IRC | 03:20 | |
| *** vnogin has quit IRC | 03:20 | |
| *** armaan has quit IRC | 03:29 | |
| *** lbragstad has joined #openstack-ansible | 03:31 | |
| *** udesale has joined #openstack-ansible | 03:48 | |
| *** agrebennikov has joined #openstack-ansible | 03:51 | |
| *** woodard has quit IRC | 03:53 | |
| *** lbragstad has quit IRC | 03:54 | |
| *** woodard has joined #openstack-ansible | 03:54 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: Set enable_dvr when router_distributed is enabled https://review.openstack.org/517187 | 03:57 |
| *** woodard has quit IRC | 03:58 | |
| *** germs has quit IRC | 04:02 | |
| *** kiennt26 has quit IRC | 04:04 | |
| *** sxc731 has quit IRC | 04:04 | |
| *** poopcat has quit IRC | 04:04 | |
| *** yifei has joined #openstack-ansible | 04:14 | |
| *** chas has joined #openstack-ansible | 04:28 | |
| *** yifei has quit IRC | 04:28 | |
| *** chas has quit IRC | 04:32 | |
| *** agrebennikov has quit IRC | 04:37 | |
| *** nshetty has joined #openstack-ansible | 05:01 | |
| *** yifei has joined #openstack-ansible | 05:05 | |
| * csmart is heading to Sydney Summit \o/ | 05:05 | |
| *** yifei has quit IRC | 05:09 | |
| *** chhavi has joined #openstack-ansible | 05:24 | |
| *** chhavi__ has joined #openstack-ansible | 05:24 | |
| *** oanson has quit IRC | 05:34 | |
| *** oanson has joined #openstack-ansible | 05:34 | |
| *** sxc731 has joined #openstack-ansible | 05:38 | |
| *** bruvik_ has joined #openstack-ansible | 06:05 | |
| *** chas has joined #openstack-ansible | 06:28 | |
| *** hachi has joined #openstack-ansible | 06:31 | |
| *** chas has quit IRC | 06:33 | |
| *** nshetty is now known as nshetty|brb | 06:44 | |
| *** nshetty|brb is now known as nshetty | 06:53 | |
| *** threestrands has quit IRC | 06:53 | |
| *** chas has joined #openstack-ansible | 06:54 | |
| *** thorst has joined #openstack-ansible | 06:54 | |
| *** kiennt26 has joined #openstack-ansible | 06:54 | |
| *** mwe1 has quit IRC | 07:01 | |
| *** sxc731 has quit IRC | 07:03 | |
| *** sxc731 has joined #openstack-ansible | 07:04 | |
| *** mwe1 has joined #openstack-ansible | 07:07 | |
| *** Oku_OS-away is now known as Oku_OS | 07:13 | |
| *** sxc731 has quit IRC | 07:16 | |
| *** hachi__ has joined #openstack-ansible | 07:17 | |
| *** hachi has quit IRC | 07:18 | |
| *** jvidal has joined #openstack-ansible | 07:19 | |
| *** nshetty is now known as nshetty|lunch | 07:21 | |
| *** chhavi has quit IRC | 07:22 | |
| *** sxc731 has joined #openstack-ansible | 07:22 | |
| *** chhavi has joined #openstack-ansible | 07:22 | |
| *** Miouge has joined #openstack-ansible | 07:24 | |
| *** pcaruana has joined #openstack-ansible | 07:24 | |
| *** vnogin has joined #openstack-ansible | 07:32 | |
| *** eumel8 has joined #openstack-ansible | 07:34 | |
| *** armaan has joined #openstack-ansible | 07:35 | |
| *** vnogin has quit IRC | 07:36 | |
| *** armaan has quit IRC | 07:48 | |
| *** hachi__ has quit IRC | 07:51 | |
| *** vnogin has joined #openstack-ansible | 07:56 | |
| *** hachi has joined #openstack-ansible | 07:56 | |
| *** vnogin has quit IRC | 08:00 | |
| *** armaan has joined #openstack-ansible | 08:11 | |
| *** gkadam has joined #openstack-ansible | 08:17 | |
| *** sxc731 has quit IRC | 08:17 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer stable/newton: Change developer_mode branch to newton-eol https://review.openstack.org/517397 | 08:23 |
| evrardjp | odyssey4me: are you here? | 08:27 |
| evrardjp | may I rebase https://review.openstack.org/#/c/512635/14 ? | 08:27 |
| *** ajmaidak has quit IRC | 08:32 | |
| jafeha | good morning | 08:32 |
| eumel8 | good morning | 08:37 |
| odyssey4me | evrardjp I am. Why would you want to rebase it? | 08:37 |
| evrardjp | because of the merge conflicts? | 08:37 |
| odyssey4me | oh I see that now :/ | 08:38 |
| odyssey4me | sure, if you like | 08:38 |
| evrardjp | it's you or me, just wanted to know :p | 08:38 |
| *** nshetty|lunch is now known as nshetty | 08:39 | |
| evrardjp | odyssey4me: for this one: https://review.openstack.org/#/c/512217/2 | 08:39 |
| evrardjp | you want to wait a little this way you backport this + the flip switch? | 08:40 |
| evrardjp | for ara gates | 08:40 |
| evrardjp | or we can do separate backports | 08:40 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum stable/newton: Change developer_mode branch to newton-eol https://review.openstack.org/517409 | 08:42 |
| odyssey4me | evrardjp that one has nothing to do with the ARA gates, unless I'm misunderstanding what you mean? | 08:43 |
| evrardjp | you wanted to add a flip switch in ara gates to always generate report right? | 08:43 |
| evrardjp | so https://review.openstack.org/#/c/512217/2 is kinda opposite to the end goal | 08:44 |
| evrardjp | or have I misunderstood? | 08:44 |
| evrardjp | I will vote on the inclusion of those, and we'll see later. | 08:44 |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat stable/newton: Change developer_mode branch to newton-eol https://review.openstack.org/517399 | 08:47 |
| *** vnogin has joined #openstack-ansible | 08:48 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance stable/newton: Change developer_mode branch to newton-eol https://review.openstack.org/517398 | 08:50 |
| eumel8 | evrardjp, odyssey4me: maybe you have some time today to talk about project doc translation in OSA | 08:50 |
| *** vnogin has quit IRC | 08:52 | |
| evrardjp | eumel8: will you be at the summit? | 08:59 |
| *** ajmaidak has joined #openstack-ansible | 09:00 | |
| evrardjp | eumel8: but yes I am ok talking about that | 09:00 |
| openstackgerrit | Merged openstack/openstack-ansible-os_sahara stable/newton: Change developer_mode branch to newton-eol https://review.openstack.org/517401 | 09:00 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_gnocchi stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/516644 | 09:01 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_ceilometer stable/newton: Initial OSA zuul v3 role jobs https://review.openstack.org/513999 | 09:06 |
| *** pylot has quit IRC | 09:07 | |
| *** pylot has joined #openstack-ansible | 09:08 | |
| eumel8 | evrardjp: no, I'm not there. | 09:09 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_gnocchi stable/newton: Use new gnocchi repo URL https://review.openstack.org/517584 | 09:10 |
| *** vnogin has joined #openstack-ansible | 09:10 | |
| evrardjp | eumel8: when are you available for this conversation? I am ok anytime. | 09:11 |
| evrardjp | I can only translate in french and dutch though | 09:11 |
| evrardjp | and my aptitudes at writing are kinda limited. | 09:11 |
| evrardjp | :p | 09:11 |
| evrardjp | but if it's around the technical side of how to do it, I am all ears | 09:12 |
| eumel8 | evrardjp: yeah, you have to join the French language team ;) | 09:13 |
| *** David_Orange has joined #openstack-ansible | 09:13 | |
| *** ajmaidak has quit IRC | 09:13 | |
| evrardjp | :) | 09:14 |
| evrardjp | Why not! | 09:14 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_gnocchi stable/newton: Move to use new upstream Gnocchi https://review.openstack.org/517586 | 09:14 |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/pike: Use zuul v3 src location to install ARA https://review.openstack.org/516917 | 09:14 |
| *** sxc731 has joined #openstack-ansible | 09:14 | |
| eumel8 | evrardjp: seriously. OSA has the most progress in project documemtation and it's a good candidate for setup translation for that. | 09:15 |
| evrardjp | eumel8: that's great to hear | 09:16 |
| evrardjp | what should we do? | 09:16 |
| evrardjp | let's make it happen | 09:16 |
| *** vnogin has quit IRC | 09:16 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_almanach stable/pike: Updated from OpenStack Ansible Tests https://review.openstack.org/517083 | 09:17 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_aodh stable/pike: Updated from OpenStack Ansible Tests https://review.openstack.org/516285 | 09:17 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_monasca stable/pike: Updated from OpenStack Ansible Tests https://review.openstack.org/516299 | 09:18 |
| eumel8 | evrardjp: I'm not shure. I think it's nearly here: https://github.com/openstack-infra/openstack-zuul-jobs/blob/master/zuul.d/project-templates.yaml#L583 | 09:18 |
| openstackgerrit | Merged openstack/ansible-hardening master: Remove unified queue from gate pipeline https://review.openstack.org/514204 | 09:18 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_swift stable/pike: Updated from OpenStack Ansible Tests https://review.openstack.org/516305 | 09:18 |
| *** pbandark has joined #openstack-ansible | 09:18 | |
| evrardjp | eumel8: ok | 09:18 |
| evrardjp | I will have a look at the jobs in question | 09:18 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_gnocchi stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/516644 | 09:19 |
| eumel8 | second part is here as an example: https://github.com/openstack/openstack-manuals/blob/master/.zuul.yaml#L111 | 09:20 |
| *** SmearedBeard has joined #openstack-ansible | 09:21 | |
| evrardjp | I was following the first one, and it brings me to a shell script named upstream_translation_update.sh | 09:26 |
| evrardjp | I still have to reverse engineer what we need to do | 09:26 |
| evrardjp | :p | 09:26 |
| evrardjp | eumel8: but it looks like we could maybe re-define something in our integrated repo | 09:27 |
| evrardjp | adding the openstack-manuals-build-translation project template | 09:27 |
| evrardjp | or re-implement something similar | 09:28 |
| evrardjp | eumel8: so we need something similar to https://github.com/openstack/openstack-manuals/blob/master/tox.ini#L58-L72 and call that into our integrated repo | 09:30 |
| evrardjp | that sounds reasonable to do | 09:30 |
| evrardjp | I will have a look with odyssey4me but I can try to draft that. | 09:30 |
| odyssey4me | evrardjp are you looking to implement translation of our docs? | 09:32 |
| evrardjp | that sounds more complex than just having https://github.com/openstack/openstack-doc-tools | 09:33 |
| evrardjp | yeah | 09:33 |
| evrardjp | Evaluating what's the cost for it at least | 09:33 |
| odyssey4me | awesome - could be interesting | 09:33 |
| odyssey4me | I do wonder what the demand is for that, but you'd be the best person to know. ;) | 09:33 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/516642 | 09:34 |
| odyssey4me | Even if there is no demand, if it's easy enough to implement and maintain then it opens avenues to new audiences so I think we should do it. | 09:34 |
| evrardjp | odyssey4me: well I heard french ppl asking last year, but it was quite limited in scope, so right now it's more evaluating what's possible as a quick win | 09:34 |
| evrardjp | I could do the french translation real quick : | 09:34 |
| evrardjp | :p | 09:34 |
| evrardjp | but let's evaluate first what would be the cost of it | 09:35 |
| andymccr | babelfish will do it automatically | 09:35 |
| evrardjp | because it would seems it would be a tangled mess :p | 09:36 |
| andymccr | :D | 09:36 |
| evrardjp | not aware of that, but with a name like that I'd not be surprised :p | 09:36 |
| evrardjp | google translate has a nice api too. Hum Hum. | 09:37 |
| odyssey4me | how can you call yourself a netizen, and a nerd, and not know about babelfish? | 09:37 |
| evrardjp | odyssey4me: never used it. | 09:37 |
| odyssey4me | before google translate, there was babelfish | 09:37 |
| evrardjp | yeah yahoo thing or something | 09:37 |
| evrardjp | kinda _heard_ about it | 09:38 |
| *** sxc731 has quit IRC | 09:38 | |
| *** electrofelix has joined #openstack-ansible | 09:38 | |
| evrardjp | never used it! | 09:38 |
| evrardjp | I think the best invention in terms of translation work is urban dictionary but don't tell eumel8 :p | 09:38 |
| odyssey4me | Yahoo! Babel Fish was a free Web-based multilingual translation application. In May 2012 it was replaced by Bing Translator, to which queries were redirected. | 09:38 |
| Neptu_ | evrardjp: Hej, and good morning. I have problems with the ssl certificates of keystone is any way to dissable the ssl on keystone while deploying?? | 09:39 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate stable/pike: Ensure tests are testing pike, not master https://review.openstack.org/517591 | 09:39 |
| evrardjp | Neptu_: yes probably, but you should probably not :p | 09:39 |
| *** udesale__ has joined #openstack-ansible | 09:39 | |
| evrardjp | is there no var from that, forcing http? | 09:39 |
| evrardjp | odyssey4me: yeah I googled it :) | 09:39 |
| evrardjp | it's weird how yahoo got ripped off all his bits nowadays | 09:40 |
| Neptu_ | evrardjp: hahaha ive been looking into the lxc container i i not see any configuration on uswig or nignx for dissable I will start looking at the role see if i find something there | 09:40 |
| evrardjp | Neptu_: yes. that's where you'd find it. And in the integrated repo, probably on the haproxy side | 09:41 |
| evrardjp | (group vars then) | 09:41 |
| Neptu_ | is a bit bizarre the way this is been done on keystone I never seen no one implementing ssl on the application rather than create a ssl on nginx... | 09:41 |
| evrardjp | uwsgi is a server, not an app. | 09:41 |
| *** udesale has quit IRC | 09:41 | |
| evrardjp | well it's an app server but let's not talk about that :) | 09:42 |
| Neptu_ | yep i know but normally is the server that provides the encrypted channel not the app | 09:42 |
| Neptu_ | ... | 09:42 |
| odyssey4me | Neptu_ SSL is terminated on the load balancer, not the back-end service | 09:42 |
| evrardjp | Neptu_: there is no "normally" imo, it all depends on the usage. I think it's smarter security wise to be closer to your app, but I can understand the constraints that go with it | 09:42 |
| *** hw_wutianwei has quit IRC | 09:43 | |
| evrardjp | odyssey4me: including for horizon and keystone? I don't remember nowadays | 09:43 |
| odyssey4me | this is largely because openstack clients historically have been a bit bad at properly catering to SSL connectivity | 09:43 |
| andymccr | i think keystone does it in nginx | 09:43 |
| odyssey4me | keystone and horizon *can* do it, but don't by default | 09:43 |
| odyssey4me | not since newton | 09:43 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate master: Allow role defaults to determine the branch tested https://review.openstack.org/517593 | 09:43 |
| andymccr | makes sense | 09:44 |
| *** drifterza has joined #openstack-ansible | 09:44 | |
| evrardjp | terminating everything at the same place is SO MUCH easier. | 09:44 |
| andymccr | although even if we did, its more a historical change in that we used to use mod_wsgi which meant the conf went in the apache/nginx conf, now that weve moved to nginx + uwsgi we could push the ssl bits to uwsgi conf | 09:44 |
| odyssey4me | the work to improve the story for PCI compliance by implementing a per control-plane host SSL endpoint is something we discussed at the PTG and jmccrory evrardjp and logan- (I think) were going to work on that | 09:44 |
| drifterza | Greetings all | 09:45 |
| evrardjp | odyssey4me: yes, we need to evaluate the solutions with the fast router and re-use logan-'s haproxy roles and improvements | 09:45 |
| evrardjp | it was kinda relegated to second citizen for now | 09:45 |
| odyssey4me | andymccr do you have a bit of time to see why https://review.openstack.org/517402 is failing on trusty ? | 09:46 |
| andymccr | sure lemme look | 09:46 |
| odyssey4me | I think it's a transient error, but it's happened twice in a row now. | 09:46 |
| odyssey4me | thanks - appreciate it | 09:46 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate stable/pike: Allow role defaults to determine the branch tested https://review.openstack.org/517591 | 09:53 |
| *** armaan has quit IRC | 09:58 | |
| Neptu_ | evrardjp: hehehe is paranoid thinking but in security is never paranoid enough.... been working with some CA company here in stockholm and definetively they do things not like in a webshop | 09:59 |
| *** ajmaidak has joined #openstack-ansible | 09:59 | |
| Neptu_ | odyssey4me: is interesting to listen to that because I do not see nginx have any option on certificates or uwisg defining any type of ssl or certificate | 10:00 |
| Neptu_ | odyssey4me: I wonder if there is an extra external layer outside the lxc-keystone container | 10:00 |
| evrardjp | Neptu_: haproxy node | 10:00 |
| odyssey4me | Neptu_ yes, as mentioned just a few lines ago in this - ssl is terminated at the load balancer | 10:00 |
| Neptu_ | aha | 10:01 |
| Neptu_ | sorry i did not understand you properly... ok ok so now i understand how you do it i was too focused on the keystone lxc container install | 10:01 |
| Neptu_ | thanks a lot i think that will do it | 10:01 |
| *** kiennt26 has quit IRC | 10:03 | |
| Neptu_ | evrardjp: odyssey4me how does it sound a contribution to hook up certbot an get electronic frontier foundation certificates by default.... is this something you would like to see on the code base? | 10:04 |
| evrardjp | Neptu_: yesterday I said I am not against it :) | 10:05 |
| evrardjp | you should probably use external_lb_vip_address as the fqdn for the domain | 10:05 |
| *** armaan has joined #openstack-ansible | 10:05 | |
| Neptu_ | evrardjp: ok i will start trying to figure out how to navigate this into the openstack burocracy | 10:05 |
| Neptu_ | something to do this weekend maybe on personal bases | 10:05 |
| evrardjp | Neptu_: if you write code it's gonna be easy to then use it | 10:05 |
| Neptu_ | im a coder | 10:05 |
| Neptu_ | and an beer drinker | 10:06 |
| evrardjp | sounds good! | 10:06 |
| *** jiaopengju has quit IRC | 10:06 | |
| evrardjp | Is there a way to re-use an external role? | 10:06 |
| Neptu_ | probably | 10:06 |
| evrardjp | Maybe use that, and make sure the deployment node gets the certificate, then distribute it. | 10:06 |
| Neptu_ | i take a look | 10:06 |
| evrardjp | ok | 10:06 |
| evrardjp | thanks | 10:06 |
| *** rpittau has joined #openstack-ansible | 10:09 | |
| odyssey4me | evrardjp did you see that 2.4 now has a certificate module | 10:10 |
| odyssey4me | https://docs.ansible.com/ansible/2.4/openssl_certificate_module.html | 10:11 |
| odyssey4me | also https://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=6&cad=rja&uact=8&ved=0ahUKEwiny_qOlaLXAhWGD8AKHf3JDS8QFghHMAU&url=https%3A%2F%2Fdocs.ansible.com%2Fansible%2F2.4%2Fopenssl_csr_module.html&usg=AOvVaw3tQY3KQJNRucCDt2HF8FJf | 10:11 |
| odyssey4me | argh | 10:11 |
| odyssey4me | https://docs.ansible.com/ansible/2.4/openssl_csr_module.html | 10:11 |
| evrardjp | oh no I didn't know | 10:12 |
| evrardjp | with acme! | 10:12 |
| evrardjp | that's great | 10:13 |
| evrardjp | we can probably provide an example playbook somewhere. | 10:14 |
| odyssey4me | well, we can likely remove our cert generation stuff with that | 10:15 |
| evrardjp | yeah | 10:15 |
| *** David_Orange has quit IRC | 10:15 | |
| evrardjp | we can probably ignore all that certificates in the future, and let the users do what they want | 10:15 |
| evrardjp | we should just propagate certs from the deploy as we discussed a few times in the past | 10:15 |
| evrardjp | simplifications! | 10:15 |
| odyssey4me | yup | 10:16 |
| *** udesale__ has quit IRC | 10:16 | |
| mardim | odyssey4me, evrardjp andymccr hello | 10:16 |
| odyssey4me | o/ | 10:16 |
| *** udesale has joined #openstack-ansible | 10:16 | |
| mardim | is it possible to help me with this https://review.openstack.org/#/c/517259/5/vars/main.yml@36 | 10:17 |
| mardim | ? | 10:17 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum stable/newton: Change developer_mode branch to newton-eol https://review.openstack.org/517409 | 10:17 |
| mardim | because I cannot see where the problem is | 10:17 |
| mardim | thanks | 10:17 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_zaqar stable/ocata: Initial Zuul v3 role jobs https://review.openstack.org/514299 | 10:22 |
| evrardjp | mardim: hello! | 10:25 |
| *** vnogin has joined #openstack-ansible | 10:25 | |
| mardim | evrardjp, hi can you please take a look on this piece of code because ansible gives me error and I do not know why | 10:26 |
| mardim | evrardjp, https://review.openstack.org/#/c/517259/5/vars/main.yml@36 | 10:26 |
| evrardjp | yes I am having a look | 10:26 |
| mardim | evrardjp, Thanks a lot | 10:26 |
| evrardjp | mardim: neutron-openvswitch-agent isn't enabled for nsh? | 10:29 |
| *** udesale__ has joined #openstack-ansible | 10:32 | |
| mardim | evrardjp, You mean if openvswitch-agent is supporting nsh protocol ? | 10:32 |
| evrardjp | if nsh is used, shouldn't neutron-openvswitch-agent be enabled too? | 10:32 |
| evrardjp | is there a reason why the two wouldn't be linked? | 10:32 |
| evrardjp | let me do a full review there :) | 10:33 |
| *** vnogin has quit IRC | 10:33 | |
| mardim | evrardjp, there is a nested if there | 10:33 |
| mardim | evrardjp, If the user wants to enable nsh support and wants also openvswitch-agent | 10:33 |
| mardim | evrardjp, then it will work | 10:34 |
| *** udesale has quit IRC | 10:34 | |
| mardim | evrardjp, the ansible will install openvswitch agent and ovs-nsh | 10:34 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_almanach stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/516636 | 10:34 |
| evrardjp | thanks for the contribution! | 10:35 |
| evrardjp | mardim: I understood that. | 10:35 |
| evrardjp | let me review it. | 10:35 |
| evrardjp | let's discuss through the review :) | 10:35 |
| mardim | evrardjp, sure :) | 10:35 |
| evrardjp | but thanks for the ping I have missed the review and all | 10:35 |
| evrardjp | so don't hesitate to ping us here :D | 10:36 |
| andymccr | mardim: i think the problem could be that if the group doestn exist it'll fail | 10:36 |
| mardim | andymccr, But the group exists because I linked it to this patch https://review.openstack.org/#/c/517269/ | 10:37 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_barbican stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/516639 | 10:37 |
| mardim | andymccr, Which means that the gates will take into account both patches when they do the build or ? | 10:37 |
| andymccr | mardim: it does - but the role tests dont use the same inventory system | 10:37 |
| andymccr | they use a much simpler system: https://github.com/openstack/openstack-ansible-os_neutron/blob/master/tests/lxb_inventory | 10:38 |
| odyssey4me | evrardjp with regards to https://review.openstack.org/#/c/515397/ - the only reason the pip_install role is doing the repo management is because we needed to cover the containers... if we just split the repo management out into its own role (or use an external role) then we would not have to refactor the openstack_hosts role | 10:38 |
| mardim | andymccr, So are you proposing to add the openvswitch_nsh group to that file ? | 10:39 |
| *** sxc731 has joined #openstack-ansible | 10:39 | |
| odyssey4me | I don't think it makes sense to refactor that role (which targets *hosts*) - it makes better sense IMO to use a playbook or role just to do repo management. | 10:39 |
| odyssey4me | evrardjp nvm, I've commented in review | 10:40 |
| andymccr | mardim: hmm so either that or test that the group exists first - im not sure which is better. i guess in the context of a full build you would have the env.d file present so integrated would work - but in the context of running the role perhaps testing it exists is a better approach | 10:41 |
| mardim | andymccr, hmm IMHO here the better aproach is to add the openvswitch_nsh group to the test file as we are doing to the other groups so we can follow the notion which already exists what do you think ? | 10:44 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_barbican stable/pike: Allow role defaults to determine the branch tested https://review.openstack.org/517601 | 10:44 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_barbican master: Allow role defaults to determine the branch tested https://review.openstack.org/517603 | 10:45 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_almanach stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/517085 | 10:45 |
| Neptu_ | the gate or aio config deploys the haproxy?? | 10:45 |
| andymccr | mardim: yeah looking at the others that seems to be whats been done - so that sounds good - just NB there are 3 inventory files (one for different scenarios) so check in tests and ensure you update each of them | 10:45 |
| andymccr | actually theres like 5 now :P | 10:45 |
| mardim | andymccr, hahah yes I will do that thanks a lot for reviewing it :) | 10:46 |
| odyssey4me | Neptu_ yes, haproxy is optionally deployed (you can use an external LB if you want to) - that's set by allocating a set of hosts to the right group, as per all the other services | 10:46 |
| *** hachi has quit IRC | 10:47 | |
| Neptu_ | odyssey4me: ok found it is not on a lxc container is on the host machine... | 10:48 |
| *** hachi has joined #openstack-ansible | 10:48 | |
| odyssey4me | Neptu_ in an AIO, by default, yes | 10:48 |
| *** vnogin has joined #openstack-ansible | 10:52 | |
| evrardjp | odyssey4me: if you see *hosts* as everything you can run stuff on, it becomes easier to understand :) | 10:53 |
| evrardjp | (ansible term basically) | 10:53 |
| evrardjp | if you see it as _bare metal_ nodes only that becomes a problem indeed | 10:54 |
| odyssey4me | evrardjp that entire role was designed to execute against hosts (things that host containers or vm's) | 10:54 |
| Neptu_ | odyssey4me: thanks found it now i need to reconfigure this manually, gonna take a while | 10:54 |
| odyssey4me | evrardjp I don't think we should perpetuate the multifunction nature of the role - the repo management should be its own role | 10:54 |
| evrardjp | odyssey4me: well I checked the whole role, and there are many overlaps. so I moved all the bare metal things into an include. It's not that hard | 10:54 |
| *** udesale__ has quit IRC | 10:54 | |
| odyssey4me | overlaps from that to what? | 10:55 |
| evrardjp | we do many things to configure hosts in many roles, if we could just prepare them for running our roles that would be good enough | 10:55 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ironic stable/ocata: Initial Zuul v3 role jobs https://review.openstack.org/514289 | 10:55 |
| evrardjp | it's a different approach | 10:55 |
| evrardjp | we can indeed move the repo management outside its role to another role if need be | 10:56 |
| evrardjp | but we still need to prepare the nodes for our stuff, and I think this role is in a good place for that | 10:56 |
| *** hachi has quit IRC | 10:57 | |
| evrardjp | I am doing simple refactors, one step at a time | 10:57 |
| odyssey4me | evrardjp ok, then perhaps I need to see what the big picture is here... could you express what you're aiming at in a spec or something? | 11:00 |
| *** nshetty has quit IRC | 11:01 | |
| *** nshetty has joined #openstack-ansible | 11:01 | |
| *** mrtenio has quit IRC | 11:06 | |
| *** hachi has joined #openstack-ansible | 11:07 | |
| *** sxc731 has quit IRC | 11:09 | |
| *** jwitko has quit IRC | 11:13 | |
| *** sxc731 has joined #openstack-ansible | 11:13 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/515082 | 11:16 |
| andymccr | i cant recreate that swift failure. that is pretty random | 11:17 |
| gunix | andymccr: what swift failure? the one i was QQing about? | 11:18 |
| odyssey4me | andymccr thanks for trying | 11:19 |
| odyssey4me | it seemed odd it was only one of the containers, not all of them | 11:19 |
| *** smatzek has joined #openstack-ansible | 11:20 | |
| andymccr | yeah and a diff one each time | 11:22 |
| andymccr | gunix: nope - an issue with trusty and older newton branch! | 11:22 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/515076 | 11:26 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh master: Allow role defaults to determine the branch tested https://review.openstack.org/517606 | 11:31 |
| *** stuartgr has joined #openstack-ansible | 11:34 | |
| *** hachi has quit IRC | 11:37 | |
| gunix | andymccr: any ideea why i can't find the create backup button in horizon for cinder volumes? backup works from cli | 11:38 |
| *** nshetty is now known as nshetty|afk | 11:41 | |
| Miouge | I get lots of “The error was: OSError: [Errno 12] Cannot allocate memory” white running the lxc_container_create task in the setup-hosts playbook. Servers have 512GB of RAM so, that should be plenty? | 11:44 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_zaqar stable/ocata: Initial Zuul v3 role jobs https://review.openstack.org/514299 | 11:44 |
| *** ajmaidak has quit IRC | 11:48 | |
| *** sxc731 has quit IRC | 11:55 | |
| mhayden | happy friday, all | 12:07 |
| *** ajmaidak has joined #openstack-ansible | 12:08 | |
| Neptu_ | odyssey4me: https://pastebin.com/Ydrft6qq Ive got the keystone internal ip attached here | 12:08 |
| Neptu_ | i wonder if haproxy is not enought | 12:08 |
| Neptu_ | to reach keystone | 12:09 |
| odyssey4me | Neptu_ is there a question in there somewhere? | 12:09 |
| Neptu_ | yes why keystone is returning me a json with information with internal ips?? | 12:09 |
| odyssey4me | Neptu_ probably because the auth information you gave was for an admin | 12:10 |
| Neptu_ | ok if i change the role of the user | 12:10 |
| Neptu_ | thanks thanks thanks! | 12:10 |
| Neptu_ | Neptu_: i changed the role of the user to heat user and still i get thejson with internal ips | 12:15 |
| evrardjp | Miouge: yes, that's far more than what's needed | 12:17 |
| evrardjp | mhayden: morning :) | 12:17 |
| *** nshetty|afk is now known as nshetty | 12:17 | |
| odyssey4me | Cores, there are some more zuul jobs PR's that need review: https://review.openstack.org/#/q/project:%22%255Eopenstack/(monitorstack%257Copenstack-ansible%257Cansible-).*%22+status:open+(topic:zuulv3-jobs+OR+topic:zuulv3-jobs-pike+OR+topic:zuulv3-jobs-ocata+OR+topic:zuulv3-jobs-newton)+label:Verified%253E%253D1%252Czuul | 12:18 |
| mhayden | has anyone seen gertty put "Held: 1 (F12)" in the status bar before? | 12:18 |
| mhayden | ah, nevermind, figured it out | 12:19 |
| mhayden | punch F12, then ! to unmark it as Held | 12:19 |
| *** thorst has quit IRC | 12:20 | |
| *** thorst has joined #openstack-ansible | 12:20 | |
| odyssey4me | there are quite a few other patches to resolve various gating issues in flight which I'd appreciate reviews for: https://review.openstack.org/#/q/project:%22%255Eopenstack/(monitorstack%257Copenstack-ansible%257Cansible-).*%22+owner:jesse-pretorius+status:open+label:Verified%253E%253D1%252Czuul | 12:20 |
| mhayden | could someone put a +2 back onto https://review.openstack.org/501263 ? | 12:20 |
| mhayden | odyssey4me: gandering in progress | 12:20 |
| odyssey4me | mhayden it says it cannot merge | 12:20 |
| mhayden | oh fantastic | 12:21 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift stable/pike: Suppress rsync warnings w/command https://review.openstack.org/501263 | 12:21 |
| odyssey4me | meh, fixed, and +2+w | 12:21 |
| mhayden | oh wow, the Zuulv3 stuff finally merged for swift in pike | 12:21 |
| mhayden | thank goodness | 12:21 |
| * mhayden was getting gray hairs | 12:21 | |
| odyssey4me | yep, we had to make centos non-voting to get that done though | 12:21 |
| odyssey4me | centos is now the most badly behaved gate ;) | 12:22 |
| mhayden | the functional tests break randomly | 12:22 |
| mhayden | it doesn't seem to be memory related | 12:22 |
| * odyssey4me is speaking generally - not just for the swift repo | 12:22 | |
| odyssey4me | bbiab | 12:23 |
| *** sxc731 has joined #openstack-ansible | 12:25 | |
| Miouge | evrardjp: No clue as to why but running lxc-containers-create.yml manually rather than via setup-hosts.yml the error (Cannot allocate memory) does not show up :D | 12:30 |
| evrardjp | mhayden: what's "held" for in gertty? | 12:32 |
| evrardjp | I still don't know :p | 12:32 |
| *** David_Orange has joined #openstack-ansible | 12:34 | |
| evrardjp | odyssey4me: hey I see you're busy with the http://logs.openstack.org/89/514289/4/check/openstack-ansible-functional-ubuntu-xenial/3da17c1/job-output.txt.gz#_2017-11-03_11_46_44_987571 | 12:40 |
| evrardjp | ironic + ceph thing | 12:40 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/pike: Set the nova-lxd test to voting https://review.openstack.org/517618 | 12:40 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/pike: Set the nova-lxd test to voting https://review.openstack.org/517618 | 12:41 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/pike: Set the nova-lxd test to voting https://review.openstack.org/517618 | 12:41 |
| Miouge | Solved the “Cannot allocate memory” error: the deployment host had too little RAM (2GB). Maybe something worth mentioning in the overview-requirements section of the deploy-guide? | 12:42 |
| odyssey4me | evrardjp I did that part, now it's failing on some other obscure error | 12:42 |
| evrardjp | Miouge: what's the size of your environment? | 12:43 |
| evrardjp | odyssey4me: yeah I want to help that's why | 12:43 |
| evrardjp | it's the recursion issue or something | 12:43 |
| evrardjp | I've seen that to some other places, so if you find out why, that would be great. | 12:43 |
| Miouge | evrardjp: 27 servers total. I run OSA from a VM, so I was maybe too conservative with RAM allocation | 12:44 |
| evrardjp | I've worked around it in the past by using different thing | 12:44 |
| odyssey4me | evrardjp I'm hands off that one now - not sure why that recursion issue is happening, and I'm tired of those patches now - need a change of scenery. | 12:44 |
| evrardjp | odyssey4me: Oh i know the feeling | 12:44 |
| evrardjp | :) | 12:44 |
| evrardjp | but you've been far more patient than me | 12:45 |
| evrardjp | :p | 12:45 |
| odyssey4me | ceilometer I'm leaving to cloudnull | 12:45 |
| evrardjp | :) | 12:45 |
| odyssey4me | it appears to be some sort of service issue there, which means actually setting something up | 12:45 |
| evrardjp | odyssey4me: he told yesterday he had some kind of instances spinned up right? | 12:45 |
| odyssey4me | almanach is a head scratcher - it's failing because it doesn't have gcc to install the pip package, and the pip_install role is in meta-deps, but it doesn't run | 12:46 |
| evrardjp | spinned up, right* ? | 12:46 |
| odyssey4me | if you could peek at https://review.openstack.org/512729 and see if you can spot something, that'd be nice | 12:46 |
| evrardjp | is it in tests arr? | 12:46 |
| odyssey4me | although at this stage I'm wondering if we should contact fdegir to find out if that's being used/maintained at all | 12:46 |
| odyssey4me | evrardjp yep it is, it runs for other roles - just not this one | 12:47 |
| evrardjp | yeah ok | 12:47 |
| *** gkadam has quit IRC | 12:47 | |
| odyssey4me | zaqar also appears broken due to some sort of internals, and I'm wondering whether that's used any more either - we should maybe reach out to kong or flwang to find out whether they plan to maintain it or not | 12:49 |
| odyssey4me | and the monasca roles are all busted | 12:50 |
| odyssey4me | I haven't looked at what the failures are though | 12:50 |
| andymccr | gonna boot 5 of these trusty nodes to test this swift thing - it failed again. on service module too that is super weird | 12:51 |
| odyssey4me | thanks andymccr | 12:52 |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift stable/pike: Updated from OpenStack Ansible Tests https://review.openstack.org/516305 | 12:55 |
| evrardjp | odyssey4me: I think I know the issue, but I will confirm in a sec | 12:56 |
| odyssey4me | yeah, aodh is working again: https://review.openstack.org/515076 | 12:56 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_almanach master: Initial OSA zuul v3 role jobs https://review.openstack.org/512729 | 12:57 |
| evrardjp | odyssey4me: should fix the issue ^ | 12:57 |
| odyssey4me | lol, you gotta be kidding me | 12:57 |
| evrardjp | :p | 12:57 |
| odyssey4me | hahaha - I clearly was tired when comparing it with other roles last night | 12:58 |
| evrardjp | yaml matters! | 12:58 |
| *** drifterza has quit IRC | 12:58 | |
| evrardjp | you did an amazing job on all those things | 12:58 |
| odyssey4me | ok, I'll do the same to the other branches | 12:58 |
| evrardjp | I will have a look at the ironic one maybe? | 12:59 |
| evrardjp | while you're busy on all the millions of others :p | 12:59 |
| odyssey4me | many of the fixes were due to a lack of maintenance - we do changes in other roles, but don't keep these according to the same patterns | 13:00 |
| *** sxc731 has quit IRC | 13:00 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_almanach stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/517085 | 13:02 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_almanach stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/516636 | 13:05 |
| *** Dinesh_Bhor has quit IRC | 13:08 | |
| mhayden | evrardjp: apparently a patch gets marked as "held" when you make an offline change that conflicts with something in gerrit when you sync again | 13:10 |
| *** smatzek has quit IRC | 13:11 | |
| evrardjp | oh | 13:11 |
| evrardjp | you got the small popup warning and then it "holds" the thing? | 13:11 |
| *** udesale has joined #openstack-ansible | 13:11 | |
| evrardjp | when I have that I quit gerrit, delete all my workspaces and reopen gerrit. | 13:11 |
| *** sxc731 has joined #openstack-ansible | 13:11 | |
| evrardjp | :D | 13:12 |
| *** jafeha has quit IRC | 13:12 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_almanach master: Initial OSA zuul v3 role jobs https://review.openstack.org/512729 | 13:17 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_almanach stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/516636 | 13:18 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_almanach stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/517085 | 13:18 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Updated role maturity matrix https://review.openstack.org/517621 | 13:20 |
| Bico_Fino | morning. | 13:20 |
| evrardjp | morning | 13:22 |
| Neptu_ | I did not managed to solve the problem of keystone dumping internal ips on me.... I think i will modify the haproxy role and deploy again openstack see how far i get... this is a time consumming issue | 13:24 |
| Bico_Fino | Finished yesterday the deploy osa + ovs, but got an error when I tried to run ovs-vsctl add-br br-provider | 13:24 |
| Neptu_ | evrardjp: do you know if there is any problems with openstack beein deployed on epic processors rather than xeon?? | 13:25 |
| openstackgerrit | Merged openstack/ansible-hardening master: [Docs] Specify STIG version in branches https://review.openstack.org/517311 | 13:29 |
| *** udesale has quit IRC | 13:30 | |
| *** lbragstad has joined #openstack-ansible | 13:31 | |
| odyssey4me | Neptu_ are your querying the public endpoint for the catalogue? what is it giving you and what are you expecting that's different from what it gives you? | 13:34 |
| odyssey4me | are you expecting a public IP, or a dns name or what? | 13:35 |
| odyssey4me | also, what is the external_lb_address which you've got configured? | 13:35 |
| *** hachi has joined #openstack-ansible | 13:36 | |
| mhayden | ooh, i love this role maturity chart doodad from evrardjp | 13:39 |
| mhayden | i need some role maturity stuff for ansible-hardening :) | 13:39 |
| evrardjp | didn't I create a patch for that? | 13:40 |
| evrardjp | Darn I thought I did | 13:40 |
| mhayden | evrardjp: i can do dat | 13:40 |
| evrardjp | oh yeah unknown | 13:40 |
| evrardjp | I will do it no worries | 13:41 |
| mhayden | evrardjp: imma fight you for it | 13:41 |
| * mhayden scurries | 13:41 | |
| evrardjp | I don't want to fight and I don't know what scurries mean | 13:41 |
| openstackgerrit | Major Hayden proposed openstack/ansible-hardening master: [WIP] I patched faster than evrardjp https://review.openstack.org/517622 | 13:42 |
| * mhayden cackles | 13:42 | |
| evrardjp | hihi | 13:42 |
| openstackgerrit | Major Hayden proposed openstack/ansible-hardening master: Add role maturity metadata to ansible-hardening https://review.openstack.org/517622 | 13:43 |
| evrardjp | still clicking it | 13:43 |
| evrardjp | :p | 13:43 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/ansible-hardening master: Role maturity https://review.openstack.org/517623 | 13:43 |
| evrardjp | darn it's gerrit that's slow | 13:43 |
| mhayden | hahaha | 13:43 |
| evrardjp | did you see what I did there? :p | 13:43 |
| evrardjp | ok cancelling mine. | 13:43 |
| mhayden | hahaha | 13:44 |
| mhayden | well played sir | 13:44 |
| mhayden | well played indeed | 13:44 |
| evrardjp | :) | 13:44 |
| evrardjp | I'll update yours to Pike and we're good? | 13:44 |
| mhayden | although i don't think i'm allowed to vouch for the 'maturity' of anything with my sense of humor | 13:44 |
| *** smatzek has joined #openstack-ansible | 13:44 | |
| mhayden | or "sense of humour" for my european friends | 13:44 |
| evrardjp | Darn what will we do then? | 13:44 |
| evrardjp | I am not better | 13:44 |
| * evrardjp cackles | 13:44 | |
| evrardjp | :D | 13:44 |
| evrardjp | mhayden: I will patch to Pike, ok? | 13:45 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/ansible-hardening master: Add role maturity metadata to ansible-hardening https://review.openstack.org/517622 | 13:45 |
| *** smatzek has quit IRC | 13:45 | |
| evrardjp | that's gonna be weird to have it backported and see created_during pike in the newton branch :p | 13:45 |
| evrardjp | but it is! | 13:46 |
| mhayden | evrardjp: hah, sure | 13:46 |
| *** smatzek has joined #openstack-ansible | 13:46 | |
| evrardjp | that's what we did effectively | 13:46 |
| mhayden | evrardjp: i thought the "created_during" was when the role was made | 13:46 |
| evrardjp | mmm | 13:47 |
| evrardjp | that's a good question | 13:47 |
| mhayden | the os_nova one says mitaka | 13:47 |
| evrardjp | yeah not juno | 13:47 |
| evrardjp | let me check | 13:47 |
| evrardjp | because this one was manual | 13:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_almanach master: Initial OSA zuul v3 role jobs https://review.openstack.org/512729 | 13:47 |
| odyssey4me | most of the roles were actually created in kilo, not mitaka - but the repositories were created in mitaka | 13:48 |
| evrardjp | yes it was based on what was tagged. | 13:48 |
| mhayden | oh yeah, with the IRR split | 13:48 |
| mhayden | feels like eons ago | 13:48 |
| evrardjp | yeah | 13:48 |
| *** jvidal has quit IRC | 13:48 | |
| evrardjp | so yes hardening should be liberty | 13:48 |
| evrardjp | because it was just a rename | 13:49 |
| evrardjp | my bad | 13:49 |
| evrardjp | let me roll it back | 13:49 |
| mhayden | evrardjp: seems fine to me either way | 13:49 |
| *** David_Orange has left #openstack-ansible | 13:49 | |
| mhayden | i just wanted to be accurate and match the others ;) | 13:49 |
| evrardjp | you're right | 13:49 |
| evrardjp | let's do it consistently | 13:49 |
| *** jwitko_ has joined #openstack-ansible | 13:49 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/ansible-hardening master: Add role maturity metadata to ansible-hardening https://review.openstack.org/517622 | 13:50 |
| *** jwitko_ is now known as jwitko | 13:50 | |
| * mhayden is perusing odyssey4me's patches | 13:50 | |
| *** jafeha has joined #openstack-ansible | 13:51 | |
| *** ivveh- has quit IRC | 13:51 | |
| evrardjp | odyssey4me: yes indeed | 13:51 |
| evrardjp | I remember when they were in | 13:51 |
| evrardjp | kilo timeframe... | 13:52 |
| evrardjp | time flies | 13:52 |
| *** sxc731 has quit IRC | 13:54 | |
| mhayden | odyssey4me: here's one for ya -> https://review.openstack.org/517622 ;) | 13:54 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_almanach stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/517085 | 13:54 |
| odyssey4me | mhayden awful nice of evrardjp to forget that role, given its importance | 13:56 |
| mhayden | odyssey4me: i cried a lot | 13:57 |
| mhayden | but i'm okay now | 13:57 |
| evrardjp | but ... | 13:57 |
| evrardjp | I am sorry! | 13:57 |
| evrardjp | openstack-ansible.* didn't match hardening | 13:57 |
| evrardjp | I know I have to update my things. | 13:57 |
| evrardjp | things == thinking mostly. | 13:58 |
| evrardjp | mhayden: what doesn't kill you makes you stronger! | 13:58 |
| cloudnull | mornings | 13:58 |
| evrardjp | cloudnull: morning :) | 13:59 |
| admin0 | \o | 13:59 |
| odyssey4me | evrardjp you need to be more welcoming to other more general ansible roles | 13:59 |
| odyssey4me | I mean, with your regex, you'd be missing logan- and your own external roles too ;) | 13:59 |
| odyssey4me | :p | 13:59 |
| evrardjp | odyssey4me: oh yes I forgot keepalived | 13:59 |
| evrardjp | darn. | 13:59 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_octavia master: Fixing a few nits in release notes https://review.openstack.org/517625 | 13:59 |
| cloudnull | odyssey4me: evrardjp: so ceilometer... I cant seem to reproduce that issue w/ trusty in my lab. It seems like a miserable race condition. | 13:59 |
| cloudnull | odyssey4me: do you know if we can backport the log collection things to newotn | 14:00 |
| cloudnull | **newton ? | 14:00 |
| mhayden | odyssey4me: I SEE WHAT YOU DID THERE ON THE MATURITY PATCH | 14:00 |
| mhayden | (one day it will be complete, i promise) | 14:00 |
| admin0 | caps alert !! | 14:00 |
| mhayden | admin0: caps lock is cruise control for cool | 14:00 |
| mhayden | that's what d34dh0r53 taught me | 14:00 |
| odyssey4me | ah, race conditions - love them so | 14:00 |
| evrardjp | mhayden: LOL | 14:00 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_almanach stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/516636 | 14:00 |
| evrardjp | ;) | 14:00 |
| evrardjp | mhayden: on top of that the "complete" word was proposed by odyssey4me himself. | 14:01 |
| mhayden | perhaps we need a YAML file with developer maturity ratings from 0-10 | 14:01 |
| admin0 | well then ..carry on .. | 14:01 |
| admin0 | :D | 14:01 |
| odyssey4me | cloudnull yeah, seeing that it all appears to be working now - lemme get on that | 14:01 |
| odyssey4me | (the log collection stuff) | 14:01 |
| cloudnull | odyssey4me: its terrible, 5 test envs today and yesterday, all passing. | 14:01 |
| evrardjp | mhayden: May I go to negatives please? | 14:01 |
| odyssey4me | cloudnull the later branches are also failing, so perhaps you can see something there? | 14:01 |
| evrardjp | mhayden: oh you meant from 0 to -10. Good! | 14:02 |
| evrardjp | https://slack-imgs.com/?c=1&url=http%3A%2F%2Friffsy.com%2Fimage%2FrpCx_tinygif.gif | 14:02 |
| mhayden | depends on whether we used a signed integer or not | 14:03 |
| cloudnull | yea its seems to be working w/ newton on xenial | 14:03 |
| mhayden | a negative number could mean that a developer is so mature that they looped back around | 14:03 |
| cloudnull | I'll begin looking into the ocata failures. | 14:03 |
| cloudnull | -https://review.openstack.org/#/c/516640/ | 14:03 |
| evrardjp | mhayden: I knew that at some point you'd be talking to buffer overflows. | 14:03 |
| evrardjp | We are closer everyday. | 14:04 |
| cloudnull | it'll give me something to do on the plane | 14:04 |
| eumel8 | evrardjp: if you can confirm, this bug should be fixed https://bugs.launchpad.net/openstack-ansible/+bug/1727779 | 14:04 |
| openstack | Launchpad bug 1727779 in openstack-ansible "[docs] Bad redirects on all the deploy guides" [Critical,In progress] - Assigned to Jean-Philippe Evrard (jean-philippe-evrard) | 14:04 |
| evrardjp | I'll have a look quick | 14:05 |
| evrardjp | yeah | 14:05 |
| evrardjp | eumel8: moved it to done. Thanks! | 14:06 |
| eumel8 | thx :) | 14:06 |
| evrardjp | Apparently I forgot to put the bug in my fixes... Sorry! | 14:06 |
| evrardjp | darn my local test just oomed | 14:07 |
| evrardjp | 30 minutes lost. | 14:09 |
| *** smatzek_ has joined #openstack-ansible | 14:11 | |
| *** smatzek has quit IRC | 14:11 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests master: Hide yum-config-manager output https://review.openstack.org/514835 | 14:12 |
| mhayden | yay | 14:13 |
| evrardjp | woot | 14:13 |
| odyssey4me | mhayden you gonna bp that one too? | 14:16 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/pike: Ensure tests repo is in required repositories https://review.openstack.org/517628 | 14:17 |
| *** agrebennikov has joined #openstack-ansible | 14:18 | |
| mardim | andymccr, Quick question | 14:20 |
| mardim | andymccr, How I am include the controll node in the openvswitch_nsh group ? | 14:20 |
| mardim | can I* | 14:21 |
| mardim | andymccr, In the neutron role tests | 14:21 |
| mardim | ? | 14:21 |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/516639 | 14:21 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_almanach master: Updated from OpenStack Ansible Tests https://review.openstack.org/514692 | 14:21 |
| andymccr | mardim: are you creating a new container? | 14:22 |
| mardim | andymccr, No | 14:22 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_monasca master: Updated from OpenStack Ansible Tests https://review.openstack.org/514695 | 14:22 |
| openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi stable/newton: Move to use new upstream Gnocchi https://review.openstack.org/517586 | 14:22 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_monasca-agent master: Updated from OpenStack Ansible Tests https://review.openstack.org/514696 | 14:22 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_monasca-ui master: Updated from OpenStack Ansible Tests https://review.openstack.org/514697 | 14:22 |
| mardim | andymccr, I mean how I represent in the tests the is metal: true | 14:23 |
| andymccr | mardim: the exitsing hosts are listed - https://github.com/openstack/openstack-ansible-os_neutron/blob/master/tests/lxb_inventory in the [all] section | 14:23 |
| andymccr | mardim: oh you don't, thats just localhost | 14:23 |
| andymccr | just add localhost to the appropriate group | 14:23 |
| andymccr | you'll see its listed as a [physical_host] | 14:23 |
| evrardjp | mardim: you can give another inventory for your role test if you need to. | 14:23 |
| mardim | andymccr, Ah ok and why we have agents1 and agents2? | 14:23 |
| mardim | andymccr, I am palnning to create one:) | 14:23 |
| mardim | planning | 14:23 |
| evrardjp | or re-use the same inventory and just override the variable to switch the usage of nsh on | 14:24 |
| andymccr | mardim: i imagine its for testing 2 agents work together | 14:24 |
| andymccr | we do that for keystone - we test ha keystone for example | 14:24 |
| *** woodard has joined #openstack-ansible | 14:24 | |
| openstackgerrit | Adrien Cunin proposed openstack/openstack-ansible-ops master: Added cleanup-venvs.yml to help remove older venvs https://review.openstack.org/517630 | 14:24 |
| *** mwe1 has quit IRC | 14:25 | |
| Adri2000 | evrardjp: ^ followup to my question yesterday (or the day before... :)) | 14:25 |
| mardim | evrardjp, andymccr I am planning to create one new inventory because I have the openvswitch_nsh group | 14:25 |
| mardim | evrardjp, andymccr Is that ok ? | 14:26 |
| evrardjp | Adri2000: that's a RADICAL approach. Should work though. You can also use the find module, but I'm not gonna complain :p | 14:26 |
| *** mwe1 has joined #openstack-ansible | 14:26 | |
| *** Brew has joined #openstack-ansible | 14:27 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican stable/pike: Allow role defaults to determine the branch tested https://review.openstack.org/517601 | 14:27 |
| andymccr | mardim: sure - that should be fine - it wont get called unless we create a new scenario though - just fyi :) and also its worth noting you'll still need that group in the other inventories because of the failure you're seeing | 14:27 |
| Adri2000 | evrardjp: ahah, tested on a real deployment this morning. was a bit nervous on the first run but well, no one complained until now so it looks like no bad side effect :p | 14:27 |
| evrardjp | Let me review that quick | 14:27 |
| Adri2000 | no rush, but thanks :) | 14:28 |
| mardim | andymccr, What do you mean I need to create a test inventry for all the roles? | 14:28 |
| mardim | andymccr, nova ,keystone ,e.t.c. ? | 14:28 |
| *** hachi has quit IRC | 14:28 | |
| andymccr | mardim: no i mean you'll need to add that group into the existing inventories | 14:28 |
| *** hachi has joined #openstack-ansible | 14:29 | |
| andymccr | and i mean having a new inventory in there wont do anything unless we create a test scenario that uses that inventory | 14:29 |
| andymccr | for example, the default test uses the lxb_inventory to test linuxbridge, and the calico test uses the calico inventory. but if you just create a new inventory then no test is actually using that inventory | 14:29 |
| mardim | andymccr, hmm can I just add this group to existing inventories and just create an variable file which will include the ovs_nsh_support variable ? | 14:29 |
| andymccr | mardim: ahh i thought you wanted to create a new inventory file | 14:30 |
| mardim | andymccr, not neceserelly that was just a thought | 14:30 |
| *** weezS has joined #openstack-ansible | 14:30 | |
| mardim | andymccr, Still trying to understand how the test are working | 14:30 |
| *** smatzek_ is now known as smatzek | 14:30 | |
| mardim | tests* | 14:30 |
| *** woodard has quit IRC | 14:31 | |
| odyssey4me | evrardjp heh, the almanach tests are now working :) | 14:31 |
| odyssey4me | tyvm for the help there | 14:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/ocata: Move log collection to 'post' stage https://review.openstack.org/517631 | 14:31 |
| evrardjp | odyssey4me: thank for the work! | 14:32 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon stable/newton: Change developer_mode branch to newton-eol https://review.openstack.org/517410 | 14:33 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone stable/newton: Change developer_mode branch to newton-eol https://review.openstack.org/517400 | 14:33 |
| *** nshetty has quit IRC | 14:34 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican master: Allow role defaults to determine the branch tested https://review.openstack.org/517603 | 14:34 |
| *** hachi has quit IRC | 14:35 | |
| evrardjp | wow gerrit starts to hate us | 14:37 |
| odyssey4me | why so? | 14:37 |
| evrardjp | got 2 error 500 in a few minutes | 14:38 |
| evrardjp | "us" was in general, not only "us" us. | 14:38 |
| evrardjp | I am pretty sure you understood me odyssey4me :) | 14:38 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/ocata: Move log collection to 'post' stage https://review.openstack.org/517631 | 14:39 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/515076 | 14:40 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/516642 | 14:41 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate master: Allow role defaults to determine the branch tested https://review.openstack.org/517593 | 14:41 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron stable/newton: Change developer_mode branch to newton-eol https://review.openstack.org/517405 | 14:42 |
| *** esberglu has joined #openstack-ansible | 14:45 | |
| openstackgerrit | Merged openstack/ansible-hardening master: Add role maturity metadata to ansible-hardening https://review.openstack.org/517622 | 14:45 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/newton: Move log collection to 'post' stage https://review.openstack.org/517635 | 14:45 |
| odyssey4me | cloudnull https://review.openstack.org/#/q/I5ab31cc141804cbcb6d8721032a79d0c724fc0a2 should do the trick | 14:46 |
| *** sc` has joined #openstack-ansible | 14:48 | |
| evrardjp | trust there! it looks good but let's go forward with the post stage ^ | 14:48 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/515082 | 14:50 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh master: Allow role defaults to determine the branch tested https://review.openstack.org/517606 | 14:50 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh stable/newton: Change developer_mode branch to newton-eol https://review.openstack.org/517396 | 14:50 |
| *** esberglu has quit IRC | 14:51 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Updated role maturity matrix https://review.openstack.org/517621 | 14:51 |
| *** esberglu has joined #openstack-ansible | 14:51 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517637 | 14:52 |
| openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/516644 | 14:53 |
| mardim | andymccr, If I will create a new variable file only how am i telling to the tests to use that variable file | 14:57 |
| mardim | andymccr, sorry for bothering you bytheway | 14:57 |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum stable/newton: Change developer_mode branch to newton-eol https://review.openstack.org/517409 | 14:57 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate stable/pike: Allow role defaults to determine the branch tested https://review.openstack.org/517591 | 14:58 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517637 | 15:02 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/pike: Update test image for nova-lxd https://review.openstack.org/517376 | 15:04 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517637 | 15:07 |
| evrardjp | odyssey4me: wow. https://review.openstack.org/#/c/517621/ | 15:10 |
| evrardjp | see the jobs running | 15:10 |
| odyssey4me | evrardjp your thoughts please - we can do https://review.openstack.org/517637 for newton and ocata, but it will require tox.ini changes in every repo to make it work... the major benefit is that it uses the pre-cloned git sources, thus making tests more reliable... the question is whether it's worth it for newton/ocata | 15:10 |
| evrardjp | for N definitely not worth it, but if we do it for O, why not doing it for N too... | 15:11 |
| evrardjp | let me think about it in more details | 15:11 |
| sc` | i've looked around, but can't find definitive documentation: how does ansible deal with service user roles? | 15:11 |
| odyssey4me | evrardjp I guess it comes down to whether we're seeing enough git clone failures in N/O to make a change like that which will cause quite a bit of churn | 15:12 |
| odyssey4me | and yes for N it's quite late, but for O it may be worth it | 15:12 |
| odyssey4me | sc` not sure I understand the question? ansible runs tasks with information given to it | 15:13 |
| odyssey4me | evrardjp how does that role maturity matrix data get updated? | 15:13 |
| sc` | i'm probably not phrasing it correctly | 15:13 |
| evrardjp | odyssey4me: that was stupid of me to bother you with the jobs there, it's because it's a doc change. | 15:13 |
| evrardjp | silly me. | 15:13 |
| evrardjp | I think for O it's worth it indeed. | 15:14 |
| odyssey4me | evrardjp oh do you need a job for that? | 15:14 |
| sc` | the mysql users are what i'm looking for specifically, like what grants they get and so on | 15:14 |
| odyssey4me | sc` is this for ansible generally, or for openstack-ansible specifically? | 15:14 |
| sc` | vanilla install docs say all privileges. my cookbooks go with least privilege | 15:14 |
| evrardjp | odyssey4me: I was talking about a different topic already :p | 15:14 |
| evrardjp | odyssey4me: I think your patch is worth it for Ocata | 15:14 |
| sc` | odyssey4me: just trying to sanity check my own project against others | 15:15 |
| odyssey4me | evrardjp I'll forgive you because it's friday | 15:15 |
| *** askb has quit IRC | 15:15 | |
| evrardjp | odyssey4me: I'm speaking terribly all the days of the week, you know it :p | 15:15 |
| sc` | sigh. too early | 15:15 |
| sc` | not enough caffeine | 15:15 |
| odyssey4me | sc` in OSA we implement a single mysql db user per project (ie cinder, etc), and a single rabbitmq user/vhost per project (ie cinder, etc) | 15:16 |
| evrardjp | odyssey4me: but I think we should at least have a patch depends-on this patch to see if it works fine | 15:16 |
| sc` | yeah. mysql is totally the wrong thing to be asking about. one moment | 15:16 |
| sc` | it's sleepy and i'm early | 15:16 |
| odyssey4me | sc` then each service has a service account in keystone too | 15:16 |
| evrardjp | odyssey4me: the advantage is that we'll have easier backports in the future, right? | 15:16 |
| evrardjp | and all the branches are behaving the same way in regards to zuul | 15:16 |
| *** askb has joined #openstack-ansible | 15:17 | |
| * odyssey4me wonders where in the docs mhayden hid his thesis on OSA security | 15:17 | |
| evrardjp | which is a good conveneince | 15:17 |
| * mhayden wats | 15:17 | |
| *** askb has quit IRC | 15:17 | |
| armaan | folks, fyi magnum/Ocata works but magun/pike doesn't. No idea why Kubernetes API never comes up -- as of now. | 15:17 |
| sc` | it's the keystone service account role that i'm looking at. i should probably just pull down a copy of osa and build a representative deployment locally | 15:17 |
| odyssey4me | evrardjp the main thing I'm actually looking to do with the patch is use the git clones zuul puts there... it's a secondary benefit that depends-on will work | 15:17 |
| evrardjp | yeah but not a small benefit | 15:18 |
| odyssey4me | sc` this may be handy? https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/app-security.html | 15:18 |
| evrardjp | I like consistency, so I'd be enclined to give it a +2. Better to ask the community that's still in O for that. So let's add jmccrory and logan- on those patches for review, and let them doing the +w if they are confident. | 15:19 |
| sc` | sort of, but it just gives the same generic stuff that my project might: least privilege, mainly | 15:19 |
| odyssey4me | evrardjp that maturity matrix sure is pretty :) | 15:19 |
| evrardjp | Could you give an example of the changes needed in the Ocata repo for a repo like os_nova ? | 15:19 |
| odyssey4me | evrardjp for now let's see if it works, which is why I did a -w | 15:19 |
| evrardjp | odyssey4me: we're gonna need to backport the matrix to other branches | 15:19 |
| sc` | i'm one of those crazies that still develops on chef (yes, it still exists) | 15:20 |
| evrardjp | odyssey4me: yes exactly | 15:20 |
| odyssey4me | sc` ah, I have fond memories of chef | 15:20 |
| sc` | i just want to make sure i'm not going too far off | 15:20 |
| sc` | i don't ansible often, so thought it might be quicker to ask than sit through a build | 15:20 |
| odyssey4me | evrardjp hmm, os_almanach is not complete - it's incubated, and it has ocata and pike branches so it must have been created in the newton cycle | 15:21 |
| evrardjp | mmm | 15:23 |
| evrardjp | it has no tests? | 15:23 |
| evrardjp | odyssey4me: it has functional test | 15:24 |
| evrardjp | so according to the definition it's not incubated anymore | 15:24 |
| evrardjp | therefore it moved to complete | 15:24 |
| odyssey4me | I don't think that https://github.com/openstack/openstack-ansible-os_almanach/blob/master/tests/test-functional-almanach.yml#L25-L35 qualifies as a functional test | 15:25 |
| evrardjp | for the creation branch, that could be a mistake, I think they are all based on the same logic, which was depending on tag and branches. | 15:25 |
| evrardjp | odyssey4me: I guess it's all a question of definition | 15:25 |
| sc` | my conumdrum is that the official installation docs for pike conflict with devstack which conflict with my own cookbooks. chef follows least privilege, but devstack grants both admin and service roles | 15:25 |
| evrardjp | for now I took the definition of "If you have something" | 15:26 |
| evrardjp | which some services don't even have! | 15:26 |
| sc` | i don't have nearly enough hardware to deploy representative puppet and ansible deployments for reference | 15:26 |
| odyssey4me | sc` sure, devstack is for development... not production | 15:26 |
| odyssey4me | both puppet and OSA have AIO deployments available - not sure if that helps? | 15:27 |
| sc` | i... know that. never mind. it was a dumb idea asking. | 15:27 |
| evrardjp | odyssey4me: the general idea is started in this role maturity, if there are bugs, we can fix them :) | 15:27 |
| odyssey4me | evrardjp sure thing | 15:27 |
| odyssey4me | looks good so far | 15:27 |
| *** sc` has left #openstack-ansible | 15:27 | |
| Neptu_ | evrardjp: can modify the conf.d/ files for the haproxy and then run the gate script... or the gate script will overwrite my changes?? | 15:28 |
| Neptu_ | i just dont want to wait an hour discover the result | 15:28 |
| evrardjp | Neptu_: don't use the gate script | 15:28 |
| evrardjp | If you're editing stuff you should follow the regular process | 15:29 |
| Neptu_ | evrardjp: i was guessing that | 15:29 |
| evrardjp | regular process for an aio is documented | 15:29 |
| evrardjp | https://docs.openstack.org/openstack-ansible/latest/contributor/quickstart-aio.html | 15:29 |
| Neptu_ | thanks for the link | 15:29 |
| evrardjp | deploy guide for non aio: https://docs.openstack.org/project-deploy-guide/openstack-ansible/latest/ | 15:29 |
| evrardjp | Neptu_: yw | 15:29 |
| odyssey4me | Neptu_ just before https://docs.openstack.org/openstack-ansible/latest/contributor/quickstart-aio.html#rebooting-an-aio see the section about running individual playbooks | 15:29 |
| odyssey4me | gate-check-commit is meant for execution by the gate, and only once - not by humans | 15:30 |
| evrardjp | odyssey4me: I'd be happy to include another table with tempest testing. When a role gets tempest tested, we add that to its metadata, and we can generate a new table based on that. | 15:30 |
| odyssey4me | evrardjp to me, for openstack roles, that is the definitive way to test - anything else is inferior | 15:31 |
| evrardjp | yeah | 15:31 |
| Neptu_ | odyssey4me: as far as i understand the gate is a CI thingie... is it? | 15:31 |
| evrardjp | but we didn't define a baseline for now. | 15:31 |
| odyssey4me | Neptu_ yes - it's the term for the CI which acts as a 'gate' for any patches which cannot merge unless they pass the 'gate' | 15:32 |
| odyssey4me | evrardjp fair enough | 15:32 |
| Neptu_ | gate keeper | 15:32 |
| openstackgerrit | Dimitrios Markou proposed openstack/openstack-ansible-os_neutron master: [DO_NOT_MERGE] Add OvS-NSH support https://review.openstack.org/517259 | 15:32 |
| Neptu_ | ohh i need to get my dungeon keeper from gog | 15:33 |
| evrardjp | lol | 15:34 |
| evrardjp | logan-: do you think it would be possible to add ceph functional testing to the nova role? | 15:35 |
| evrardjp | else we can just make sure that a change in os_nova is tested with the integrated gate jobs | 15:36 |
| andymccr | do we have the functionality to create alias ips inside containers? | 15:36 |
| evrardjp | andymccr: I think we can do whatever we want in the container creation script if need be | 15:37 |
| evrardjp | if that needs to be permanent, I am not sure. | 15:37 |
| Neptu_ | the changes on the haproxy need to be done before or after the boostrap-aio? | 15:37 |
| Neptu_ | i think before | 15:37 |
| andymccr | evrardjp: its a random request i know :D | 15:37 |
| andymccr | actually i could just use diff ports instead of ips | 15:37 |
| odyssey4me | andymccr I think we do - we have a very flexible set of functionality for nic definitions | 15:38 |
| andymccr | odyssey4me: yeah was hoping i could do container_networsk def for a host - i'll test it | 15:38 |
| evrardjp | andymccr: with different ports you can simply use the container_networks | 15:38 |
| andymccr | im not sure it'll let me specify the same bridge twice | 15:38 |
| evrardjp | andymccr: it sounds like a terrible idea network wise | 15:39 |
| evrardjp | I'll call you mr loops . | 15:39 |
| odyssey4me | hmm, it doesn't look like https://github.com/openstack/openstack-ansible-lxc_container_create/blob/master/templates/debian-interface.cfg.j2 will allow more than one address item per interface | 15:39 |
| andymccr | evrardjp: haha no just like an alias | 15:39 |
| andymccr | like 192.168.10.10 and 192.168.10.11 on the same host | 15:40 |
| evrardjp | yeah on the same nic it would make sense | 15:40 |
| andymccr | eth0:0 eth0:1 | 15:40 |
| andymccr | w/e | 15:40 |
| evrardjp | iface :1 | 15:40 |
| evrardjp | yeah | 15:40 |
| evrardjp | post-up script or in the creation script :D | 15:40 |
| odyssey4me | heh, oh yeah - just found that myself | 15:40 |
| odyssey4me | so it should work | 15:41 |
| *** armaan has quit IRC | 15:46 | |
| *** armaan has joined #openstack-ansible | 15:46 | |
| evrardjp | odyssey4me: did you see what I said a little above with the nova role? | 15:49 |
| evrardjp | we basically merged something that broke ceph and we didn't see it. because ceph gets proper testing in the integrated gate, I was wondering if that was possible | 15:49 |
| evrardjp | a little like keystone workflow but with ... os_nova :p | 15:50 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517637 | 15:50 |
| *** armaan has quit IRC | 15:51 | |
| evrardjp | andymccr: do you remember an ansible 2.2 bug with the include_role? | 15:53 |
| evrardjp | I remember we discussed it. | 15:53 |
| evrardjp | I can't find a link though | 15:53 |
| *** smatzek has quit IRC | 15:54 | |
| odyssey4me | evrardjp it had something to do with leaking vars IIRC, I think in relation to delegation | 15:54 |
| odyssey4me | might have been a bug, but might have been us not understanding how to use it | 15:54 |
| andymccr | evrardjp: there was definitely an include bug in templating | 15:55 |
| andymccr | but i think that still exists, it was a jinja2 issue | 15:55 |
| andymccr | or rather an ansible implementation of jinja2 issue | 15:55 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron stable/pike: Move the neutron_dhcp_config back to defaults https://review.openstack.org/515952 | 15:57 |
| odyssey4me | evrardjp jmccrory keen to get nova-lxd for pike voting again? https://review.openstack.org/517618 | 15:57 |
| *** smatzek has joined #openstack-ansible | 15:57 | |
| evrardjp | odyssey4me: no doubt. | 15:57 |
| evrardjp | andymccr: yes I think we are hitting it in nova. | 15:58 |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/newton: Move log collection to 'post' stage https://review.openstack.org/517635 | 15:58 |
| andymccr | evrardjp: link? | 15:58 |
| evrardjp | links even! | 15:59 |
| evrardjp | 1) https://github.com/openstack/openstack-ansible-ceph_client/blob/master/tasks/ceph_all.yml#L27 the task that fails | 15:59 |
| evrardjp | gets included from https://github.com/openstack/openstack-ansible-os_nova/blob/stable/ocata/tasks/main.yml#L116-L127 | 15:59 |
| *** jafeha has quit IRC | 15:59 | |
| evrardjp | and 3) the stacktrace | 15:59 |
| Neptu_ | evrardjp: I do not see the haproxy conf on the aio installation, is not included? | 16:00 |
| evrardjp | (loading) | 16:00 |
| evrardjp | it is included | 16:01 |
| evrardjp | it runs haproxy-install.yml iirc | 16:01 |
| evrardjp | during a part of the process. | 16:01 |
| odyssey4me | Neptu_ it is - what're you looking for? the resulting config files, or the files that ansible uses? | 16:01 |
| evrardjp | andymccr: I think I will give you the full link without the line: http://logs.openstack.org/89/514289/4/check/openstack-ansible-functional-ubuntu-xenial/3da17c1/job-output.txt.gz | 16:02 |
| evrardjp | my browser is close to dying here. | 16:02 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_ceilometer stable/newton: Updated from OpenStack Ansible Tests https://review.openstack.org/516259 | 16:02 |
| andymccr | hahaha | 16:02 |
| andymccr | ok cool | 16:02 |
| andymccr | not sure what to do about that swift job btw | 16:02 |
| Neptu_ | odyssey4me: Im looking for having keystone without ssl therefore i udnerstand i need to modify the haproxy.yml inlcuded in the conf.d. Ive seen that file on the gate deployment but i not see it now on the aio deployment which feels strange | 16:02 |
| andymccr | i cant recreate that issue at all ive built like 10 of those now | 16:02 |
| odyssey4me | Neptu_ no, you don't need to modify conf.d or any haproxy config | 16:03 |
| odyssey4me | Neptu_ turning off ssl for endpoints is described in the 'test lab' appendix example config | 16:03 |
| odyssey4me | andymccr recheck 4 eva ? | 16:04 |
| andymccr | haha yeah it seems so :( | 16:04 |
| andymccr | while true; | 16:04 |
| Neptu_ | odyssey4me: can u please point me the documentation for this test lab? | 16:04 |
| odyssey4me | Neptu_ https://docs.openstack.org/project-deploy-guide/openstack-ansible/pike/app-config-test.html | 16:04 |
| odyssey4me | right at the bottom of the page it shows how to disable https public endpoints | 16:05 |
| Neptu_ | odyssey4me: ok i double check that one when the install is finished | 16:06 |
| *** dxiri has quit IRC | 16:06 | |
| odyssey4me | Neptu_ better to set that before the install | 16:06 |
| Neptu_ | odyssey4me: the problem i had aswell was related to get public endpoints pubished by keystone... I create new roles and used this users but i still got | 16:07 |
| Neptu_ | odyssey4me: ok i fix that before the install | 16:07 |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift stable/newton: Change developer_mode branch to newton-eol https://review.openstack.org/517402 | 16:10 |
| Neptu_ | odyssey4me: openstack_service_publicuri_proto: http that was easy... | 16:10 |
| andymccr | haha as if by magic it merges | 16:12 |
| mardim | andymccr, evrardjp I add the group to the test_inventories but I still get the same error in the zuul | 16:13 |
| Neptu_ | odyssey4me: regarding keystone giving away internal ip's is there a way to solve this?? | 16:13 |
| mardim | added* | 16:13 |
| evrardjp | andymccr: I've really thought about something that will recheck automatically all my patches until I say "no more" or it merges. | 16:15 |
| *** Oku_OS is now known as Oku_OS-away | 16:15 | |
| odyssey4me | Neptu_ you still haven't given enough information to help solve that one - are you querying the catalog, and is the user you're querying with an admin or not, and what are you getting versus what are you expecting? | 16:16 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Added cleanup-venvs.yml to help remove older venvs https://review.openstack.org/517630 | 16:16 |
| andymccr | evrardjp: haha yeah that would be good | 16:17 |
| evrardjp | lol | 16:17 |
| evrardjp | don't tempt me! | 16:17 |
| evrardjp | it takes like 50 lines max I guess. | 16:17 |
| *** smatzek has quit IRC | 16:18 | |
| *** smatzek has joined #openstack-ansible | 16:18 | |
| *** smatzek has quit IRC | 16:20 | |
| *** smatzek has joined #openstack-ansible | 16:20 | |
| *** vnogin has quit IRC | 16:21 | |
| *** armaan has joined #openstack-ansible | 16:23 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_almanach master: Initial OSA zuul v3 role jobs https://review.openstack.org/512729 | 16:24 |
| openstackgerrit | Merged openstack/openstack-ansible-os_almanach stable/ocata: Initial OSA zuul v3 role jobs https://review.openstack.org/516636 | 16:24 |
| openstackgerrit | Merged openstack/openstack-ansible-os_almanach stable/pike: Initial OSA zuul v3 role jobs https://review.openstack.org/517085 | 16:28 |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/ocata: Move log collection to 'post' stage https://review.openstack.org/517631 | 16:30 |
| *** smatzek has quit IRC | 16:31 | |
| Neptu_ | odyssey4me: ok let me finish the install and I will generate some logs for you to undestand better | 16:32 |
| Neptu_ | but i have 1h more and is weekend!! | 16:32 |
| Neptu_ | is been a long weekend im thinking eve online | 16:32 |
| odyssey4me | evrardjp good news - https://review.openstack.org/517635 is working, tested with https://review.openstack.org/513999 | 16:33 |
| evrardjp | that's nice! | 16:34 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517650 | 16:34 |
| evrardjp | I am still on the figuring out why the ceph thing | 16:34 |
| odyssey4me | testing https://review.openstack.org/517631 with https://review.openstack.org/516640 now | 16:34 |
| odyssey4me | also testing https://review.openstack.org/517637 with https://review.openstack.org/517650 now | 16:35 |
| evrardjp | I think our gates will be considered the best gates | 16:35 |
| *** smatzek has joined #openstack-ansible | 16:36 | |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-ops master: [WIP] Prevent deploying old code once leaped https://review.openstack.org/517126 | 16:36 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_aodh stable/ocata: Updated from OpenStack Ansible Tests https://review.openstack.org/517651 | 16:37 |
| evrardjp | "I like to surround myself from the best gates. I know the best gates. We have the best gates." -- world famous quote. | 16:38 |
| odyssey4me | hahaha | 16:38 |
| odyssey4me | good news - proposal bot is doing the right thing now: https://review.openstack.org/517651 | 16:39 |
| *** smatzek has quit IRC | 16:41 | |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-ops master: Prevent deploying old code once leaped https://review.openstack.org/517126 | 16:49 |
| *** idlemind has quit IRC | 16:51 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh stable/ocata: Updated from OpenStack Ansible Tests https://review.openstack.org/517651 | 16:52 |
| *** bruvik_ has quit IRC | 16:59 | |
| *** bruvik_ has joined #openstack-ansible | 16:59 | |
| *** a0x3e8 has joined #openstack-ansible | 17:05 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_octavia stable/pike: Enable V2 Octavia API (Experimental) https://review.openstack.org/515698 | 17:06 |
| *** smatzek has joined #openstack-ansible | 17:07 | |
| odyssey4me | heh, it works evrardjp https://review.openstack.org/#/c/517650/ | 17:07 |
| evrardjp | that is freaking great. | 17:08 |
| *** Smeared_Beard has joined #openstack-ansible | 17:08 | |
| evrardjp | logan-: ^ jmccrory ^ could you vote on the depends on of this patch if you agree for this kind of new features backported in stable branch? | 17:09 |
| evrardjp | I am ok with it myself | 17:09 |
| evrardjp | mgariepy: too | 17:09 |
| *** dxiri has joined #openstack-ansible | 17:09 | |
| jmccrory | does that already go back to ocata otherwise? | 17:09 |
| *** SmearedBeard has quit IRC | 17:09 | |
| evrardjp | we're gonna do ocata too in the same way | 17:10 |
| evrardjp | if I understood properly | 17:10 |
| odyssey4me | jmccrory evrardjp yeah, I'll do a port to ocata too - I was just trying it out with the worst case scenario | 17:10 |
| odyssey4me | doing one update now quickly to remove projects which don't have newton branches | 17:11 |
| odyssey4me | I can automate the patches to all the other repositories | 17:11 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517637 | 17:13 |
| Neptu_ | what is the working version for the python clients for keystone etc... v2 or v3?? | 17:13 |
| *** rpittau has quit IRC | 17:13 | |
| evrardjp | Neptu_: keystone v2 is abandoned in master. | 17:13 |
| evrardjp | abandonned* | 17:13 |
| odyssey4me | Neptu_ v2 or v3 is the keystone API version, not client version | 17:14 |
| odyssey4me | v2 is being removed in Queens (master) | 17:14 |
| evrardjp | and yes it's api version | 17:14 |
| Neptu_ | i mean | 17:14 |
| evrardjp | sorry for my abuse of language! | 17:14 |
| Neptu_ | i shoudl load the v3 or v2 rc file | 17:14 |
| evrardjp | I don't understand | 17:14 |
| *** jamesdenton has quit IRC | 17:14 | |
| Neptu_ | evrardjp: is fine, .. no worries .. you pay for beeer in vancuver | 17:14 |
| Neptu_ | :D | 17:15 |
| odyssey4me | jmccrory evrardjp if you guys think it'll be worth it then I'll propose an ocata test set too | 17:15 |
| Neptu_ | eeeeeeeeee | 17:16 |
| Neptu_ | ok very weird, very very very weird | 17:16 |
| *** chas has quit IRC | 17:17 | |
| *** chas has joined #openstack-ansible | 17:17 | |
| Neptu_ | https://pastebin.com/NLfQJmX7 | 17:18 |
| Neptu_ | i have all the endpoinds fixed with http not https | 17:18 |
| *** gouthamr has joined #openstack-ansible | 17:18 | |
| Neptu_ | i do the call with http on the browser and i get nothing | 17:19 |
| Neptu_ | same with https and works (kinda) | 17:19 |
| evrardjp | odyssey4me: yeah worth it im | 17:19 |
| evrardjp | imo | 17:19 |
| evrardjp | Neptu it's not the first time you're coming around with http and https thingies, do you run external and internal lb vip on the same address? | 17:20 |
| Neptu_ | evrardjp: donno i just runned the aio install and changed the install to make the haproxy use http | 17:21 |
| evrardjp | odyssey4me: fun stuff about the ceph bug | 17:21 |
| *** weezS has quit IRC | 17:21 | |
| Neptu_ | evrardjp: i feel lost i thought i had this under control | 17:21 |
| evrardjp | so when doing the include role and then doing the 2 levels of include inside ceph client role | 17:21 |
| Neptu_ | is slaping me on the face basically | 17:22 |
| *** chas has quit IRC | 17:22 | |
| evrardjp | the first task of the ceph client second level include fails. | 17:22 |
| evrardjp | Whatever the module is | 17:22 |
| Neptu_ | need to go now feel free to comment i will read it from home | 17:22 |
| evrardjp | Neptu_: just start from something that's standard | 17:22 |
| evrardjp | AIO should be working fine with https | 17:22 |
| evrardjp | if you need a specific certificate, modify your config, and give it your certificate | 17:23 |
| Neptu_ | evrardjp: is working monasca api is not linking self signed certs | 17:23 |
| evrardjp | if it doesn't like self signed certs, I am pretty sure it won't like http | 17:23 |
| evrardjp | :p | 17:23 |
| Neptu_ | well there is 2 problems first is the ssl layer which i can deactrivate from the haproxy | 17:23 |
| evrardjp | and maybe that's an issue to fix like allowing self signed cert | 17:23 |
| evrardjp | (--insecure option in clients or something like this) | 17:24 |
| *** gouthamr_ has joined #openstack-ansible | 17:24 | |
| Neptu_ | hehehe i was browsing the code today the insecure flag is never used on the code... | 17:24 |
| Neptu_ | so kaboom happy hour | 17:24 |
| Neptu_ | they are throwing me out fro the office now talk later | 17:25 |
| andymccr | evrardjp: not sure that is the same old issue: https://github.com/ansible/ansible/issues/23609 | 17:25 |
| andymccr | it seems to be impacting on 2.4 | 17:25 |
| evrardjp | yeah | 17:27 |
| evrardjp | same issue | 17:27 |
| *** gouthamr has quit IRC | 17:27 | |
| andymccr | that is weird though | 17:27 |
| andymccr | also seems ansible just went "nah dont want to fix that" | 17:27 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/pike: Set the nova-lxd test to voting https://review.openstack.org/517618 | 17:28 |
| evrardjp | I am thinking of removing the ceph_all thing | 17:28 |
| evrardjp | it's a fun thing, because it's not on the task itself that it fails, it's when you're invoking the content of the module | 17:32 |
| evrardjp | so not on the strategy | 17:33 |
| evrardjp | very interesting | 17:33 |
| jmccrory | odyssey4me where does it checkout the correct branch of roles that are cloned by zuul? | 17:34 |
| *** gouthamr_ is now known as gouthamr | 17:36 | |
| *** lbragstad has quit IRC | 17:37 | |
| jmccrory | hmm i see. guess it's some pre-run zuul task http://logs.openstack.org/50/517650/1/check/openstack-ansible-functional-ubuntu-xenial/ee40062/job-output.txt.gz#_2017-11-03_16_37_49_031313 | 17:39 |
| evrardjp | logan-: do you know if we moved everything that include ceph_client to an include role? This way I know if I can remove some condition... Else I'd check myself | 17:43 |
| logan- | not sure about gnocchi but nova/glance/cinder are changed for sure | 17:44 |
| andymccr | logan-: on that note - could we note use the ceph_client role in ceph-ansible instead? :D | 17:44 |
| *** vnogin has joined #openstack-ansible | 17:44 | |
| andymccr | and then have integration bits for the openstack conf | 17:44 |
| logan- | andymccr: was looking at that the other day. would love to do that | 17:44 |
| *** vnogin has quit IRC | 17:45 | |
| andymccr | logan-: ok cool - i'll take some time at some point to look into that | 17:45 |
| *** vnogin has joined #openstack-ansible | 17:45 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_octavia master: Add allocation pool for octavia mgmt subnet https://review.openstack.org/515378 | 17:47 |
| openstackgerrit | Merged openstack/openstack-ansible-os_octavia master: add segmentation id option to octavia mgmt network creation task https://review.openstack.org/515035 | 17:47 |
| *** sc68cal has quit IRC | 17:48 | |
| *** germs has joined #openstack-ansible | 17:51 | |
| *** vnogin has quit IRC | 17:52 | |
| *** germs1 has joined #openstack-ansible | 17:52 | |
| *** vnogin has joined #openstack-ansible | 17:53 | |
| *** germs has quit IRC | 17:56 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_almanach master: Updated from global requirements https://review.openstack.org/517665 | 17:58 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_almanach master: Updated from global requirements https://review.openstack.org/517665 | 18:00 |
| *** vnogin has quit IRC | 18:01 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/ocata: Used cached git sources and enable depends-on https://review.openstack.org/517666 | 18:03 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/ocata: Used cached git sources and enable depends-on https://review.openstack.org/517666 | 18:05 |
| odyssey4me | jmccrory evrardjp logan- the equivalent ocata patch: https://review.openstack.org/517666 | 18:06 |
| *** sc68cal has joined #openstack-ansible | 18:06 | |
| odyssey4me | oh dear - I used the same change id so the depends-on will get complicated | 18:07 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/ocata: Used cached git sources and enable depends-on https://review.openstack.org/517667 | 18:08 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517669 | 18:09 |
| odyssey4me | mhayden jmccrory evrardjp logan- sorry, had to abandon those to prevent issues when doing the rest of them | 18:11 |
| odyssey4me | https://review.openstack.org/517669 is the same patch, just different change id | 18:11 |
| mhayden | ORLY | 18:12 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517650 | 18:12 |
| odyssey4me | yeah, sorry - it took me a moment to realise that if both the newton and ocata patches have the same change ID, then all the depends-on patches would wait for *both* to merge before they could merge | 18:13 |
| evrardjp[m] | oh yeah that's true. | 18:13 |
| odyssey4me | mhayden lol, it took less than a minute for the gate to respond to your comment in https://review.openstack.org/517665 | 18:15 |
| *** chhavi has quit IRC | 18:16 | |
| *** chhavi__ has quit IRC | 18:16 | |
| *** vnogin has joined #openstack-ansible | 18:18 | |
| mhayden | kaboom | 18:19 |
| odyssey4me | alright, lemme make nodepool earn its keep | 18:19 |
| *** vnogin has quit IRC | 18:23 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517670 | 18:25 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517670 | 18:27 |
| odyssey4me | apologies for the spam - incoming patch storm | 18:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/ansible-hardening stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517671 | 18:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-apt_package_pinning stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517672 | 18:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ceph_client stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517673 | 18:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_client stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517674 | 18:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517675 | 18:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-haproxy_server stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517676 | 18:29 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517678 | 18:29 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517679 | 18:29 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-memcached_server stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517680 | 18:29 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-openstack_hosts stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517681 | 18:29 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-apt_package_pinning stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517672 | 18:29 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-openstack_openrc stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517682 | 18:29 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517683 | 18:29 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517684 | 18:30 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517685 | 18:30 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_gnocchi stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517686 | 18:30 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517687 | 18:30 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517688 | 18:30 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ironic stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517689 | 18:30 |
| *** vnogin has joined #openstack-ansible | 18:30 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517690 | 18:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517691 | 18:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/ansible-hardening stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517671 | 18:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517692 | 18:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517694 | 18:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_sahara stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517695 | 18:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517696 | 18:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517697 | 18:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-pip_install stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517698 | 18:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-plugins stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517699 | 18:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rabbitmq_server stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517700 | 18:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_build stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517701 | 18:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_server stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517702 | 18:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rsyslog_client stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517703 | 18:33 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rsyslog_server stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517704 | 18:33 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517705 | 18:33 |
| openstackgerrit | Merged openstack/openstack-ansible-os_almanach master: Updated from global requirements https://review.openstack.org/517665 | 18:35 |
| *** vnogin has quit IRC | 18:35 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-haproxy_server stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517676 | 18:35 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-plugins stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517699 | 18:38 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517696 | 18:39 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517696 | 18:40 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517696 | 18:41 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517696 | 18:42 |
| gunix | ook so ansible 17 is for queens. i was wondering why i have random stuff missing from horizon. i got unstable :)) | 18:42 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517696 | 18:45 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517694 | 18:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517694 | 18:49 |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/pike: Ensure tests repo is in required repositories https://review.openstack.org/517628 | 18:50 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/ocata: Set the nova-lxd test to voting https://review.openstack.org/517706 | 18:50 |
| *** Smeared_Beard has quit IRC | 18:53 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517692 | 18:54 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_monasca stable/pike: Updated from OpenStack Ansible Tests https://review.openstack.org/516299 | 18:57 |
| *** woodard has joined #openstack-ansible | 18:58 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517692 | 18:58 |
| *** woodard has quit IRC | 19:03 | |
| mgariepy | anyone seen a greenfield deployment with neutron db not being updated to the latest version ? | 19:06 |
| *** stuartgr has left #openstack-ansible | 19:07 | |
| mgariepy | I had to run: "neutron-db-manage upgrade head" to fix it. | 19:08 |
| mgariepy | centos: pike 16.0.2 | 19:08 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests master: Simplify install_pkg_deps() function https://review.openstack.org/517710 | 19:09 |
| *** weezS has joined #openstack-ansible | 19:11 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517690 | 19:14 |
| *** snowman48391 has joined #openstack-ansible | 19:16 | |
| *** snowman4839 has quit IRC | 19:16 | |
| gunix | mgariepy: that is strange. | 19:16 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517683 | 19:18 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517683 | 19:18 |
| *** chas has joined #openstack-ansible | 19:18 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-openstack_openrc stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517682 | 19:19 |
| mgariepy | gunix, yes I know :) haha | 19:19 |
| mgariepy | i'll start a vm monday and test it on my own. I didn't ran the deployment on that one. just helped debugging it. | 19:20 |
| *** electrofelix has quit IRC | 19:21 | |
| *** chas has quit IRC | 19:22 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ironic stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517689 | 19:22 |
| *** vnogin has joined #openstack-ansible | 19:25 | |
| evrardjp | prepping for the plane for Sydney. please don't forget the bug triage next week! Sorry if I won't be able to attend... :'( | 19:28 |
| evrardjp | see you laterz! | 19:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517683 | 19:29 |
| odyssey4me | evrardjp just send an email indicate that it's not going to happen if no-one has volunteered | 19:30 |
| armaan | folks, need some recommendation on decreasing the memory usage by OSA containers? I suppose we can always reduce the process threads? | 19:30 |
| *** pcaruana has quit IRC | 19:32 | |
| gunix | can you run osa on deb instead of ubuntu ? | 19:32 |
| *** snowman48391 has quit IRC | 19:32 | |
| gunix | armaan: i have something for you if you run POC | 19:32 |
| gunix | meaning, test | 19:32 |
| armaan | gunix: awesome | 19:33 |
| gunix | you can limit the number of workers/service to 1 | 19:33 |
| gunix | it's how osa-aio deploys | 19:33 |
| armaan | in the template | 19:33 |
| gunix | and gets your control to use 6 GB ram | 19:33 |
| armaan | gunix: how can i do that? | 19:34 |
| gunix | armaan: https://github.com/openstack/openstack-ansible/blob/master/tests/roles/bootstrap-host/templates/user_variables.aio.yml.j2#L32-L137 | 19:35 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ironic stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517689 | 19:35 |
| gunix | armaan: adding that to user variables and running setup_openstack.yml should get ram on controller from 10 gb to 6 | 19:36 |
| armaan | gunix: man, thanks a million! This will help me a lot. :) | 19:36 |
| gunix | armaan: actually cloudnull gave me that a few days ago cause i had the same issue. don't use that in prod, though | 19:37 |
| odyssey4me | armaan every service has a max_threads var | 19:37 |
| odyssey4me | armaan depending on which release you're working with, you can use the AIO overrides as a short-cut reference, for example: https://github.com/openstack/openstack-ansible/blob/master/tests/roles/bootstrap-host/templates/user_variables.aio.yml.j2#L32-L137 | 19:38 |
| armaan | odyssey4me: I have to run atleast 50-80 multi node openstack environments which will spawn only two instances. So i guess i could use the same vars which are in aio? | 19:38 |
| odyssey4me | you can inspect the roles themselves - I wouldn't suggest tuning it that far down... you'll likely want more than a single or two threads... but you can set it down dramatically | 19:39 |
| gunix | my 2 cents: give your controller enough ressources if you have that many nodes | 19:39 |
| armaan | odyssey4me: Release: Ocata, Okay, perhaps up to 8gb per controller should be okay. | 19:40 |
| *** snowman4839 has joined #openstack-ansible | 19:40 | |
| odyssey4me | the default formula for almost all services is shown as an example here: https://github.com/openstack/openstack-ansible-os_glance/blob/master/defaults/main.yml#L158-L159 | 19:40 |
| armaan | gunix: I meant 50-80 openstack deployments. Openstack on top of Openstack. | 19:40 |
| odyssey4me | that equates to - the number of CPU's divided by two, or 16 - whichever is smaller | 19:40 |
| odyssey4me | you will also find that galera will take whatever it can get | 19:41 |
| gunix | odyssey4me: any ideea why demo user after tempest deploy can't use swift? i google before asking, found nothing :( | 19:41 |
| odyssey4me | in the AIO we tune it *way* down: https://github.com/openstack/openstack-ansible/blob/master/tests/roles/bootstrap-host/templates/user_variables.aio.yml.j2#L23-L27 | 19:41 |
| odyssey4me | gunix it should be able to, assuming you have swift deployed | 19:42 |
| gunix | nope. can't create containers, and of course can't use backups, since access is per user | 19:42 |
| logan- | i have problems running a small dev cluster on 3x16gb nodes without tuning workers | 19:42 |
| armaan | odyssey4me: Okay, currently i am using 16gb for controller and my objective is to bring it to 8 gb. I am also running 3 galera containers and 3 rabbit containers. I suppose i could reduce that number 1 container of each service? | 19:44 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517683 | 19:45 |
| armaan | logan-: How much is the RAM consumption per controller? | 19:45 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops stable/newton: Used cached git sources and enable depends-on https://review.openstack.org/517683 | 19:46 |
| *** Brew has quit IRC | 19:47 | |
| odyssey4me | armaan as much as you give it | 19:47 |
| odyssey4me | if you want to reduce ram usage, you need to tune the workers down, and you should probably also tune galera and rabbit down heavily | 19:47 |
| odyssey4me | our defaults cater for a mid-sized environment, not a small one | 19:47 |
| armaan | odyssey4me: I guess what i want to confirm is whether it is okay to have just 1 galera container and 1 rabbit container in a multi node test environment? | 19:48 |
| armaan | odyssey4me: you are right i definitely need to tune the workers but not sure about the best way to tune galera and rabbit. | 19:49 |
| odyssey4me | armaan it depends on your needs - if you do your backups well and are prepared to have outages, then yes | 19:50 |
| openstackgerrit | Merged openstack/openstack-ansible-os_octavia master: Fixing a few nits in release notes https://review.openstack.org/517625 | 19:50 |
| odyssey4me | if you want high availability, then you need 3 of each | 19:50 |
| *** SmearedBeard has joined #openstack-ansible | 19:50 | |
| armaan | HA is not a requirement at the moment. It is just for the summit workshop. | 19:51 |
| odyssey4me | armaan I've given you the galera tuning - look through the rabbit role and you'll find tuning options there too - along with some research online | 19:51 |
| odyssey4me | another option is also to actually set the containers to only be able to use an amount of RAM | 19:51 |
| odyssey4me | https://blog.codeship.com/lxc-memory-limit/ | 19:51 |
| armaan | odyssey4me: Ahh yes, just saw the galera one again, i mixed up the links. | 19:52 |
| odyssey4me | you can force the container to be limited in the ram it can use | 19:52 |
| odyssey4me | if you're doing a training thing, that's probably your best bet | 19:52 |
| *** a0x3e8 has quit IRC | 19:54 | |
| armaan | odyssey4me: it is for this one, https://www.openstack.org/summit/sydney-2017/summit-schedule/events/19911/practical-kubernetes-with-magnum-and-openstack-ansible?BackURL=https%3A%2F%2Fwww.openstack.org%2Fsummit%2Fsydney-2017%2Fsummit-schedule%2Fspeakers%2F398%23eventid%3D19911 | 19:54 |
| odyssey4me | armaan ah, nice - setting up training environments is hard | 19:54 |
| odyssey4me | the mix of consistency and low resources is quite a challenge | 19:54 |
| odyssey4me | if it's for training, start with the AIO config tuning and see how that goes | 19:55 |
| odyssey4me | you don't actually need 3 controllers at all - so you could optionally use a dedicated host for rabbit/galera and another host for the openstack bits... we've built OSA so that you can do just about anything you want, as long as you know what you want and are prepared to figur out how to do it | 19:56 |
| *** SmearedBeard has quit IRC | 19:56 | |
| *** a0x3e8 has joined #openstack-ansible | 19:56 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Update swift role SHA https://review.openstack.org/517522 | 19:57 |
| armaan | I really like your suggestion of "force the container to be limited in the ram it can use", i am gonna test that. I am using one controller and affinity set to 3 for galera and rabbit, i will set it to 1 now. | 19:57 |
| armaan | Oh yes, the flexibility provided by OSA in OpenStack space is second to none. | 19:58 |
| Neptu_ | screen -list | 20:00 |
| odyssey4me | yeah, if you have a single controller, as long as you're running newton+ you can just use a single rabbit/galera... no need to cluster if you don't need HA | 20:01 |
| odyssey4me | if you change the container cpu/ram limits, you need to shut the container down and start it up again - a reboot won't apply the changes | 20:01 |
| *** a0x3e8 has quit IRC | 20:03 | |
| *** a0x3e8 has joined #openstack-ansible | 20:06 | |
| *** hachi has joined #openstack-ansible | 20:07 | |
| armaan | roger that, let me make these changes and see what comes out of it. Thanks a lot for the extremely useful information! | 20:07 |
| *** a0x3e8 has quit IRC | 20:07 | |
| *** a0x3e8 has joined #openstack-ansible | 20:07 | |
| *** vnogin has quit IRC | 20:12 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests stable/ocata: Used cached git sources and enable depends-on https://review.openstack.org/517667 | 20:17 |
| odyssey4me | armaan good luck with the training session - hope it works well! | 20:18 |
| armaan | Thank you! Florian will conduct it, i am just setting it up. | 20:22 |
| odyssey4me | haha, no pressure :p | 20:23 |
| odyssey4me | I'm out for the weekend, or at least for tonight - cheers all! | 20:23 |
| *** smatzek has quit IRC | 20:30 | |
| *** smatzek has joined #openstack-ansible | 20:31 | |
| *** hachi has quit IRC | 20:34 | |
| gunix | odyssey4me: after giving the demo user the swiftoperator role within the demo project, i was able to create the volumebackups container wihtin the dem project, using demo user. after this, cinder volume backups started to work | 20:35 |
| *** smatzek has quit IRC | 20:35 | |
| gunix | cloudnull: are you online? | 20:37 |
| *** hachi has joined #openstack-ansible | 20:38 | |
| *** chas has joined #openstack-ansible | 20:38 | |
| gunix | cloudnull: this was too long for irc. i sent you an email from gunix@gunix.cloud | 21:15 |
| eumel8 | gunix: he's travelling - https://twitter.com/cloudnull/status/926549063700533248 | 21:19 |
| *** hachi has quit IRC | 21:19 | |
| *** thorst has quit IRC | 21:19 | |
| gunix | he doesn't have access to mail at the summit, does he | 21:20 |
| *** a0x3e8 has quit IRC | 21:25 | |
| *** a0x3e8 has joined #openstack-ansible | 21:25 | |
| gunix | andymccr: can i pls forward the mail to you, since cloudnull is at the summit? it's about swift. :D | 21:27 |
| *** a0x3e8 has quit IRC | 21:28 | |
| *** SmearedBeard has joined #openstack-ansible | 21:55 | |
| *** SmearedBeard has quit IRC | 22:01 | |
| *** Miouge has quit IRC | 22:05 | |
| *** snowman4839 has quit IRC | 22:12 | |
| *** snowman4839 has joined #openstack-ansible | 22:13 | |
| *** kmARC has joined #openstack-ansible | 22:30 | |
| *** chas has quit IRC | 22:31 | |
| *** dxiri has quit IRC | 22:52 | |
| *** eumel8 has quit IRC | 22:59 | |
| *** thorst has joined #openstack-ansible | 22:59 | |
| *** thorst has quit IRC | 23:00 | |
| *** eumel8 has joined #openstack-ansible | 23:01 | |
| *** weezS has quit IRC | 23:06 | |
| *** mwynne has joined #openstack-ansible | 23:07 | |
| *** eumel8 has quit IRC | 23:16 | |
| *** vnogin has joined #openstack-ansible | 23:27 | |
| *** gouthamr has quit IRC | 23:40 | |
| *** SmearedBeard has joined #openstack-ansible | 23:57 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
