| *** thorst has quit IRC | 00:04 | |
| *** sdake_ has quit IRC | 00:05 | |
| *** jamesdenton has joined #openstack-ansible | 00:06 | |
| *** jamesdenton has quit IRC | 00:11 | |
| *** rahuls has quit IRC | 00:18 | |
| *** markvoelker has joined #openstack-ansible | 00:20 | |
| *** markvoelker has quit IRC | 00:25 | |
| *** sdake has joined #openstack-ansible | 00:38 | |
| *** sdake has quit IRC | 00:38 | |
| *** sdake has joined #openstack-ansible | 00:38 | |
| *** thorst has joined #openstack-ansible | 00:56 | |
| *** thorst has quit IRC | 00:58 | |
| *** thorst has joined #openstack-ansible | 00:58 | |
| *** sacharya has joined #openstack-ansible | 00:59 | |
| *** saneax is now known as saneax_AFK | 01:02 | |
| *** sacharya has quit IRC | 01:04 | |
| *** thorst has quit IRC | 01:07 | |
| *** jasondotstar has quit IRC | 01:18 | |
| *** spotz_zzz is now known as spotz | 01:29 | |
| openstackgerrit | Michael Davies proposed openstack/openstack-ansible: Add an easy way to run cmds in utility container https://review.openstack.org/326200 | 01:38 |
|---|---|---|
| *** sdake has quit IRC | 01:50 | |
| *** sacharya has joined #openstack-ansible | 01:59 | |
| *** thorst has joined #openstack-ansible | 02:05 | |
| *** sdake has joined #openstack-ansible | 02:11 | |
| *** thorst has quit IRC | 02:12 | |
| *** alikins has quit IRC | 02:15 | |
| *** wadeholler has quit IRC | 02:16 | |
| *** klamath has quit IRC | 02:16 | |
| *** klamath has joined #openstack-ansible | 02:17 | |
| *** alikins has joined #openstack-ansible | 02:17 | |
| *** sdake_ has joined #openstack-ansible | 02:18 | |
| *** sdake has quit IRC | 02:20 | |
| *** markvoelker has joined #openstack-ansible | 02:21 | |
| *** klamath has quit IRC | 02:24 | |
| *** klamath has joined #openstack-ansible | 02:25 | |
| *** markvoelker has quit IRC | 02:26 | |
| *** jamielennox is now known as jamielennox|away | 02:27 | |
| *** jamielennox|away is now known as jamielennox | 02:35 | |
| *** bbmbx__ has quit IRC | 02:37 | |
| *** Drago has quit IRC | 02:53 | |
| *** thorst has joined #openstack-ansible | 03:10 | |
| *** thorst has quit IRC | 03:17 | |
| *** sdake_ has quit IRC | 03:38 | |
| *** evilt0ne has quit IRC | 03:38 | |
| *** zerda2 has joined #openstack-ansible | 03:43 | |
| *** aludwar1 has joined #openstack-ansible | 03:44 | |
| *** jamielennox is now known as jamielennox|away | 03:44 | |
| *** aludwar1 has quit IRC | 03:44 | |
| *** Drago has joined #openstack-ansible | 03:48 | |
| *** Drago has quit IRC | 03:48 | |
| *** Drago has joined #openstack-ansible | 03:48 | |
| *** daneyon_ has joined #openstack-ansible | 03:51 | |
| *** jamesdenton has joined #openstack-ansible | 03:52 | |
| *** jamesdenton has quit IRC | 03:52 | |
| *** daneyon has quit IRC | 03:54 | |
| *** jamielennox|away is now known as jamielennox | 03:59 | |
| *** daneyon has joined #openstack-ansible | 04:00 | |
| *** daneyon_ has quit IRC | 04:03 | |
| *** pcaruana has joined #openstack-ansible | 04:11 | |
| *** aludwar has quit IRC | 04:12 | |
| *** Drago has quit IRC | 04:13 | |
| *** thorst has joined #openstack-ansible | 04:15 | |
| *** psilvad has quit IRC | 04:21 | |
| *** markvoelker has joined #openstack-ansible | 04:22 | |
| *** thorst has quit IRC | 04:22 | |
| *** pcaruana has quit IRC | 04:24 | |
| *** markvoelker has quit IRC | 04:26 | |
| *** aludwar has joined #openstack-ansible | 04:28 | |
| *** pcaruana has joined #openstack-ansible | 04:34 | |
| *** sacharya has quit IRC | 04:36 | |
| errr | if I have a custom theme for horizon does the horizon playbook support deploying it, or would I need to handle that some other way? | 04:37 |
| *** sacharya has joined #openstack-ansible | 04:37 | |
| *** pcaruana has quit IRC | 04:40 | |
| *** sacharya has quit IRC | 04:42 | |
| *** thorst has joined #openstack-ansible | 05:22 | |
| *** markvoelker has joined #openstack-ansible | 05:23 | |
| *** markvoelker has quit IRC | 05:27 | |
| *** thorst has quit IRC | 05:28 | |
| *** sacharya has joined #openstack-ansible | 05:38 | |
| *** sacharya has quit IRC | 05:43 | |
| *** thorst has joined #openstack-ansible | 06:25 | |
| *** thorst has quit IRC | 06:32 | |
| *** sguduru has joined #openstack-ansible | 06:51 | |
| *** mpjetta has quit IRC | 07:05 | |
| *** chhavi has joined #openstack-ansible | 07:10 | |
| *** bsv has joined #openstack-ansible | 07:13 | |
| *** markvoelker has joined #openstack-ansible | 07:23 | |
| *** markvoelker has quit IRC | 07:28 | |
| *** mikelk has joined #openstack-ansible | 07:29 | |
| *** thorst has joined #openstack-ansible | 07:30 | |
| evrardjp | good morning everyone | 07:30 |
| evrardjp | v1k0d3n: I'm there now | 07:30 |
| evrardjp | chris_hultin: will check on msgpack-python | 07:31 |
| evrardjp | chris_hultin: I have the version 0.4.7 on mitaka, what are you looking for? | 07:32 |
| winggundamth | good morning | 07:36 |
| winggundamth | on here http://docs.openstack.org/developer/openstack-ansible/install-guide/configure-network-services.html | 07:37 |
| evrardjp | lbragstad: hope your environment is fine now | 07:37 |
| evrardjp | winggundamth: yup? | 07:37 |
| *** thorst has quit IRC | 07:37 | |
| winggundamth | it said that there are router, firewall, lbaas, vpnaas, metering and qos services | 07:37 |
| evrardjp | I don't remember qos, but yes, you can | 07:38 |
| winggundamth | but what I found in neutron role are lbaas, firewall, vpnaas, vpnaas | 07:38 |
| winggundamth | I don't see router, metering and qos anywhere? | 07:38 |
| winggundamth | also how exactly to use lbaasv2. because I tried lbaasv2 and I got ImportError: Plugin 'lbaasv2' not found error | 07:39 |
| *** saneax_AFK is now known as saneax | 07:39 | |
| evrardjp | did you follow this? http://docs.openstack.org/developer/openstack-ansible/install-guide/configure-network-services.html#deploying-lbaas-v2 | 07:40 |
| winggundamth | http://docs.openstack.org/developer/openstack-ansible/install-guide/configure-lbaas.html this doc said please use lbaasv2 | 07:40 |
| evrardjp | yes | 07:40 |
| winggundamth | but http://docs.openstack.org/developer/openstack-ansible/install-guide/configure-network-services.html#load-balancing-service-optional said use neutron_lbaas.services.loadbalancer.plugin.LoadBalancerPluginv2 | 07:40 |
| evrardjp | lbaas v1 is pretty muuch dead | 07:40 |
| winggundamth | so I a bit confuse :) | 07:40 |
| evrardjp | the doc seems fine, could you drop your user_variables somewhere? | 07:41 |
| evrardjp | all of this is in the neutron role | 07:42 |
| evrardjp | the agents part | 07:42 |
| winggundamth | yep I put it in user_variables.yml | 07:42 |
| winggundamth | my point is there are 2 docs that conflict together | 07:42 |
| evrardjp | winggundamth: the file is neutron_upstart_init | 07:43 |
| winggundamth | one said put lbaasv2 and another said put neutron_lbaas.services.loadbalancer.plugin.LoadBalancerPluginv2 in neutron_plugin_base | 07:43 |
| evrardjp | oh | 07:43 |
| winggundamth | I tried lbaasv2 but it doesn't work so I'm trying neutron_lbaas.services.loadbalancer.plugin.LoadBalancerPluginv2 right now | 07:43 |
| winggundamth | and yeah neutron_lbaas.services.loadbalancer.plugin.LoadBalancerPluginv2 is the correct one | 07:44 |
| evrardjp | yes it's the latest doc | 07:45 |
| winggundamth | I think neutron_lbaas.services.loadbalancer.plugin.LoadBalancerPluginv2 this is not consistency with the others | 07:45 |
| evrardjp | I don't know where you found the other | 07:45 |
| winggundamth | it should be lbaasv2 | 07:45 |
| winggundamth | haha I found that via google actually | 07:46 |
| evrardjp | ok | 07:46 |
| evrardjp | old doc | 07:46 |
| evrardjp | we have a problem with old doc appearing then | 07:47 |
| evrardjp | could you file a bug with the two links and mention there is a problem of consistency, please? | 07:48 |
| evrardjp | One should be deleted, we should check with the docs part of our team | 07:48 |
| evrardjp | asettle: this is gonna be for you | 07:49 |
| winggundamth | no problem :) | 07:53 |
| evrardjp | thanks winggundamth! | 07:55 |
| winggundamth | this is done https://bugs.launchpad.net/openstack-ansible/+bug/1589842 | 07:57 |
| openstack | Launchpad bug 1589842 in openstack-ansible "Old document still appears" [Undecided,New] | 07:57 |
| winggundamth | and still waiting for @Mattt to answer this question :) https://review.openstack.org/#/c/325768/ | 07:58 |
| evrardjp | I can answer for him | 08:07 |
| evrardjp | run-playbooks isn't forced to be an AIO | 08:07 |
| evrardjp | most likely it will be | 08:08 |
| evrardjp | but it's not forced | 08:08 |
| evrardjp | I don't see a problem of having that there | 08:08 |
| evrardjp | but feel free to commit otherwise | 08:08 |
| evrardjp | we'll evaluate if it's better or not :D | 08:09 |
| *** iceyao has joined #openstack-ansible | 08:14 | |
| *** sguduru has quit IRC | 08:15 | |
| *** thorst has joined #openstack-ansible | 08:35 | |
| asettle | evrardjp: what's up? what am I doing? Who do I need to kill? | 08:38 |
| asettle | I mean, doc | 08:38 |
| evrardjp | :D | 08:39 |
| evrardjp | asettle: https://bugs.launchpad.net/openstack-ansible/+bug/1589842 | 08:39 |
| openstack | Launchpad bug 1589842 in openstack-ansible "Old document still appears" [Undecided,New] | 08:39 |
| *** rcarrillocruz has joined #openstack-ansible | 08:39 | |
| rcarrillocruz | heya folks | 08:39 |
| rcarrillocruz | question | 08:39 |
| rcarrillocruz | i have a folk asking in #ansible how to use the openstack modules | 08:40 |
| asettle | evrardjp: bug for me? | 08:40 |
| rcarrillocruz | he just installed an AIO with openstack-ansible to test stuff | 08:40 |
| evrardjp | asettle: yup | 08:40 |
| rcarrillocruz | unsure if openstack-ansible installs a clouds.yaml as devstack does? | 08:40 |
| evrardjp | it's about docs: we have two files explaining how to configure lbaas | 08:40 |
| evrardjp | and ofc they aren't consistent | 08:41 |
| evrardjp | but you're reworking the docs | 08:41 |
| evrardjp | so my guess is you are cleaning up stuff | 08:41 |
| asettle | evrardjp: I'll take a squiz when I'm done with the emails :) thanks for pointing it in my direction | 08:42 |
| *** thorst has quit IRC | 08:42 | |
| evrardjp | asettle: I can delete it for you if you want, but my guess is: we probably want to study how this happened and what do we do to avoid that in the future | 08:44 |
| asettle | evrardjp: so, duplication of information is not exactly uncommon. I'd like to take a look before we simply just delete the information. | 08:44 |
| evrardjp | ok, that's what I thought :D | 08:45 |
| evrardjp | Thanks ! | 08:45 |
| asettle | No worries evrardjp :) i'll keep you in the loop today as I go through it. | 08:45 |
| *** electrofelix has joined #openstack-ansible | 09:03 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create: Test external connectivity for containers https://review.openstack.org/288909 | 09:15 |
| pjm6 | good morning all | 09:16 |
| evrardjp | morning | 09:17 |
| pjm6 | evrardjp, o/ | 09:17 |
| *** markvoelker has joined #openstack-ansible | 09:24 | |
| asettle | odyssey4me and evrardjp - do we want to even include the LBaaS v1 information? Also, the bug suggests to delete the shittier one. Ha. So I might move the LBaaS information from the network services section out to the configure-lbaas chap and just have a link ref to it in the network services section | 09:27 |
| asettle | And do we have upgrade information for LBaaS 1 to 2? | 09:28 |
| evrardjp | there is no upgrade path from 1 to 2 | 09:28 |
| asettle | Well that isn't fantastic. | 09:28 |
| evrardjp | and we should use the v2 | 09:28 |
| asettle | Is there *going* to be an upgrade path? | 09:28 |
| evrardjp | and everybody should use 2 | 09:28 |
| evrardjp | not in osa IIRC | 09:28 |
| evrardjp | I think using neutron_lbaas.services.loadbalancer.plugin.LoadBalancerPluginv2 is the correct way for neutron_plugin_base | 09:28 |
| *** markvoelker has quit IRC | 09:29 | |
| asettle | Should is the key word there evrardjp - there has been use cases for v1 before Mitaka, and the original doc tells you how to use v1. You *should* have an upgrade path. | 09:29 |
| evrardjp | but doc wise, I don't know what to do | 09:29 |
| asettle | evrardjp: that I can do. | 09:29 |
| asettle | There should be an upgrade path though. So far, the option is to manually delete load balancers, pools, and members. | 09:29 |
| asettle | And that's hardly user friendly. | 09:29 |
| asettle | odyssey4me: was there ever a plan for this? | 09:29 |
| evrardjp | that's sad we advised to use v1, but that's maybe what happened at that time because it was the best plan | 09:30 |
| evrardjp | at that time | 09:30 |
| evrardjp | now times have changed | 09:30 |
| asettle | evrardjp: yes, well, that's fine :) that happens. But there still needs to be an upgrade option because it was something that was originally used/in the doc. | 09:30 |
| asettle | Just because times have changed, you can't ignore upgrades. | 09:30 |
| evrardjp | No but we can mention it's outside the scope (if it is) | 09:31 |
| evrardjp | we should mention* | 09:31 |
| evrardjp | if it's inside the scope, we should better write code now :p | 09:31 |
| asettle | I will - definitely. But I think we should have another option (if even temporary) for this. Is there a potential upgrade path? | 09:31 |
| asettle | I'll take care of the doc now so that it reflects what we're doing. | 09:32 |
| evrardjp | that's a question for odyssey4me | 09:32 |
| evrardjp | and for the move of the doc file, I had a hard time finding the configure-lbaas document because it wasn't in the doctree | 09:32 |
| evrardjp | so I'm not sure if this needs to be adapted too | 09:33 |
| evrardjp | I completely trust you on this | 09:33 |
| asettle | evrardjp: please do :) i'll be done in a jiffy | 09:33 |
| asettle | Once I find the damn file. Stupid thing | 09:33 |
| evrardjp | find . -name configure-lbaas.rst ? | 09:33 |
| bsv | evrardjp: no, thats cheating... | 09:34 |
| odyssey4me | alextricity25 I'm tracking https://bugs.launchpad.net/neutron/+bug/1523031 waiting for the upstream fix. There's nothing we can do until there is an upstream fix. The workaround is not to use L2POP, which from Mitaka onwards we do not do. | 09:34 |
| openstack | Launchpad bug 1523031 in openstack-ansible trunk "Neighbor table entry for router missing with Linux bridge + L3HA + L2 population" [High,Confirmed] - Assigned to Jesse Pretorius (jesse-pretorius) | 09:34 |
| *** chhavi has quit IRC | 09:34 | |
| evrardjp | bsv :D | 09:34 |
| *** admin0 has joined #openstack-ansible | 09:34 | |
| odyssey4me | errr you can set a dict of files to copy from the deployment host to the horizon containers: https://github.com/openstack/openstack-ansible-os_horizon/blob/master/tasks/horizon_post_install.yml#L51-L57 | 09:36 |
| *** thorst has joined #openstack-ansible | 09:39 | |
| *** sacharya has joined #openstack-ansible | 09:40 | |
| openstackgerrit | Merged openstack/openstack-ansible: [GATE UNBLOCKER] Lock CPU map script to a release https://review.openstack.org/325578 | 09:43 |
| *** sacharya has quit IRC | 09:45 | |
| odyssey4me | rcarrillocruz sorry for the late reply - I'm still catching up on scrollback... yes we do place a clouds.yaml file down into the utility container | 09:46 |
| rcarrillocruz | oh nice | 09:46 |
| rcarrillocruz | thanks odyssey4me | 09:46 |
| *** admin0 has quit IRC | 09:46 | |
| odyssey4me | rcarrillocruz specifically in the openstack_openrc role we drop it down: https://github.com/openstack/openstack-ansible-openstack_openrc/blob/master/tasks/main.yml#L36-L44 | 09:47 |
| odyssey4me | I think that actually gets placed in more than just the utility container | 09:47 |
| *** thorst has quit IRC | 09:47 | |
| odyssey4me | rcarrillocruz yeah, in fact it should be on the host in the AIO too because the nova role will put it there: https://github.com/openstack/openstack-ansible/blob/master/playbooks/os-nova-install.yml#L263 | 09:48 |
| rcarrillocruz | providing it's on the utility host, that's fine | 09:49 |
| *** permalac has joined #openstack-ansible | 09:57 | |
| *** wadeholler has joined #openstack-ansible | 09:57 | |
| *** wadeholler has quit IRC | 09:57 | |
| *** wadeholler has joined #openstack-ansible | 09:58 | |
| *** evilt0ne has joined #openstack-ansible | 10:00 | |
| *** bsv has quit IRC | 10:03 | |
| *** johnmilton has quit IRC | 10:05 | |
| *** admin0 has joined #openstack-ansible | 10:05 | |
| winggundamth | evrardjp: so https://review.openstack.org/#/c/325768/ for this. I'm waiting for Jenkins to +1. I think someone already make a patch and he said it will run the test again automatically | 10:05 |
| evrardjp | wait a sec my browser crashed | 10:06 |
| winggundamth | haha no problem | 10:07 |
| winggundamth | that always happen when you leave it open for a day :P | 10:07 |
| evrardjp | too many tabs and I close at least 30 yesterday | 10:07 |
| evrardjp | I should abandon ff | 10:07 |
| evrardjp | it doesn't scale with my usage | 10:07 |
| winggundamth | I always clean up all the tabs few times a day so I can survive this | 10:08 |
| *** admin0 has quit IRC | 10:09 | |
| evrardjp | winggundamth: the thing is also to restart browser at regular times, which I forget | 10:12 |
| winggundamth | :) | 10:13 |
| *** admin0 has joined #openstack-ansible | 10:16 | |
| Bofu2U | morning | 10:22 |
| winggundamth | anyone has experienced on creating new domain? I have a problem that when create domain, it doesn't work correctly because there is no service account. | 10:22 |
| winggundamth | so what is the step to create domain properly? | 10:23 |
| Bofu2U | evrardjp I think I found the problem | 10:28 |
| odyssey4me | winggundamth there is nothing special that needs to be done to create a domain - you simply create one using your super-admin account | 10:33 |
| openstackgerrit | Darren Birkett proposed openstack/openstack-ansible-os_swift: remove ansible-managed line from swift.conf https://review.openstack.org/326369 | 10:33 |
| winggundamth | odyssey4me: oh sorry. I didn't tell that I'm using active directory with osa configuration | 10:34 |
| winggundamth | so it created domain automatically. but I found that after I assigned one of user in AD as admin of that domain. there is no service account and show me the error like Error: Invalid service catalog service: compute | 10:35 |
| *** admin0 has quit IRC | 10:35 | |
| winggundamth | I can logged-in via horizon but I don't know what is the next step to create all the service accounts. | 10:36 |
| Bofu2U | odyssey4me: whats the best way to take a container out of rotation, and re-create it to the point where it grabs a different IP address? | 10:38 |
| Bofu2U | As much as I admire it's aspirations, my galera container is not a neutron agent. | 10:38 |
| openstackgerrit | Darren Birkett proposed openstack/openstack-ansible: remove ansible-managed line from swift.conf https://review.openstack.org/326374 | 10:38 |
| odyssey4me | ah winggundamth now your question makes more sense - you're trying to put your default domain in ldap | 10:41 |
| odyssey4me | that is not the best of ideas | 10:41 |
| *** johnmilton has joined #openstack-ansible | 10:41 | |
| winggundamth | odyssey4me: no, I'm doing in separate domain as the doc said here http://docs.openstack.org/developer/openstack-ansible/install-guide/configure-keystone.html#implementing-ldap-or-active-directory-backends | 10:42 |
| winggundamth | so I enabled horizon_keystone_multidomain_support to True and I can logged-in let say MyCorp domain now | 10:43 |
| odyssey4me | winggundamth ok, good - but then there should be no need for a service account in that domain | 10:43 |
| odyssey4me | keystone will search the AD back-end using the credentials you provide in the back-end config | 10:44 |
| odyssey4me | as your service accounts are in a different domain, nothing should be trying to access them in the second domain | 10:44 |
| odyssey4me | unless something is not configured with a domain | 10:44 |
| evrardjp | Bofu2U: so you had an issue with your inventory | 10:45 |
| odyssey4me | are you only seeing the search for that account when browsing through horizon, or does it happen when you use the CLI too? | 10:45 |
| evrardjp | :D | 10:45 |
| winggundamth | odyssey4me: yes and seems like everything working correctly. I can see all the user list in the domain. | 10:45 |
| evrardjp | you can use scripts/inventory-manage -r <item> | 10:45 |
| evrardjp | .py | 10:45 |
| evrardjp | it will remove the item from your inventory, and you'll have a new thingy on next inventory run | 10:46 |
| Bofu2U | evrardjp so that's the fun part - technically it's not a problem with the inventory per-se | 10:46 |
| evrardjp | you'll have to recreate it | 10:46 |
| Bofu2U | the inventory shows another IP for the container | 10:46 |
| Bofu2U | for the galera container | 10:46 |
| winggundamth | odyssey4me: but when I logged-in with my AD account that I already assigned it as admin role via horizon. I'm trying to create new projects. it can created but it shows Error: Unable to set project quotas. | 10:46 |
| evrardjp | when you tell me this, it looks like an issue | 10:47 |
| odyssey4me | winggundamth ah, hrm.... | 10:47 |
| Bofu2U | evrardjp well yes, but I make it a rule not to accept blame before 7 AM :P | 10:47 |
| *** thorst has joined #openstack-ansible | 10:47 | |
| evrardjp | could you dump your openstack_user_config and your openstack_inventory.json? | 10:47 |
| *** johnmilton has quit IRC | 10:47 | |
| evrardjp | I don't blame you | 10:48 |
| winggundamth | odyssey4me: oops. it found it | 10:48 |
| Bofu2U | evrardjp yeah sec, just removed it going to see if I can see SSH commands in the neutron container as it runs | 10:48 |
| evrardjp | I blame the issue between the keyboard and chair that happened in the past :p | 10:48 |
| Bofu2U | touche | 10:48 |
| Bofu2U | PEBKAC errors are the worst. | 10:48 |
| evrardjp | our inventory never forgets, which is gonna be changed soon | 10:49 |
| Bofu2U | you're gonna make it forget? | 10:49 |
| odyssey4me | soon (tm) : | 10:49 |
| odyssey4me | :p | 10:49 |
| Bofu2U | hahah | 10:49 |
| Bofu2U | by the way, I believe I found out why this whole clusterF caused the NAT'd subnets not to work | 10:49 |
| Bofu2U | controller3, neutron agent was set as the "gateway" for the internal network. | 10:50 |
| winggundamth | odyssey4me: so the problem is it doesn't have any project/tenant. so after you created project you have to select that project first and everything work as expected now. | 10:50 |
| Bofu2U | which is the one that's misbehaving. | 10:50 |
| *** smatzek has joined #openstack-ansible | 10:52 | |
| *** thorst has quit IRC | 10:53 | |
| openstackgerrit | Darren Birkett proposed openstack/openstack-ansible-os_swift: remove ansible-managed line from swift.conf https://review.openstack.org/326369 | 10:55 |
| openstackgerrit | Darren Birkett proposed openstack/openstack-ansible: remove ansible-managed line from swift.conf https://review.openstack.org/326374 | 10:56 |
| *** chhavi has joined #openstack-ansible | 10:57 | |
| openstackgerrit | Darren Birkett proposed openstack/openstack-ansible-os_swift: Make ansible managed line in swift.conf static https://review.openstack.org/326369 | 10:57 |
| *** weshay has joined #openstack-ansible | 10:57 | |
| openstackgerrit | Darren Birkett proposed openstack/openstack-ansible: Make ansible managed line in swift.conf static https://review.openstack.org/326374 | 10:58 |
| *** shausy has joined #openstack-ansible | 10:58 | |
| *** admin0 has joined #openstack-ansible | 10:59 | |
| *** johnmilton has joined #openstack-ansible | 10:59 | |
| *** admin0 has quit IRC | 11:01 | |
| winggundamth | just wonder. there is no lbaasv2 gui on horizon yet? | 11:04 |
| mancdaz | winggundamth I think there is a plugin | 11:04 |
| winggundamth | is it install with osa yet? | 11:05 |
| mancdaz | winggundamth https://github.com/openstack/openstack-ansible-os_horizon/blob/master/tasks/horizon_post_install.yml#L59-L66 | 11:06 |
| *** wadeholler has quit IRC | 11:07 | |
| *** wadeholler has joined #openstack-ansible | 11:07 | |
| winggundamth | hmm interesting. let me recheck with horizon configuration again if it apply or not. | 11:07 |
| winggundamth | I don't see any load balance menu in horizon | 11:07 |
| mancdaz | winggundamth ping mhayden later if you're having issues - he added that stuff | 11:08 |
| winggundamth | lol it can direct access via /project/loadbalancers/ url but it can not got this error Error: Unable to retrieve pools list | 11:12 |
| odyssey4me | winggundamth yes, there is a horizon panel enabled - it should do it automatically for you | 11:12 |
| odyssey4me | https://github.com/openstack/openstack-ansible-os_horizon/blob/master/defaults/main.yml#L147 | 11:13 |
| odyssey4me | https://github.com/openstack/openstack-ansible/blob/master/playbooks/inventory/group_vars/hosts.yml#L235 | 11:13 |
| *** admin0 has joined #openstack-ansible | 11:14 | |
| odyssey4me | winggundamth so if you have neutron_plugin_base in user_variables.yml, and it contains the entry 'neutron_lbaas.services.loadbalancer.plugin.LoadBalancerPluginv2' then the panel will automatically be enabled once you next run the os-horizon-install.yml playbook | 11:14 |
| winggundamth | oh so I have to enable that individually right? | 11:14 |
| odyssey4me | winggundamth nope, it must just be in the list | 11:15 |
| winggundamth | odyssey4me: that's weird because it's not | 11:15 |
| odyssey4me | winggundamth it should be done as per http://docs.openstack.org/developer/openstack-ansible/install-guide/configure-network-services.html#deploying-lbaas-v2 | 11:15 |
| winggundamth | check /etc/horizon/local_settings.py and got 'enable_lb': False, | 11:15 |
| winggundamth | let me re deploy again | 11:16 |
| odyssey4me | winggundamth yes, that's correct - that it the old horizon panel set for lbaas v1 | 11:16 |
| winggundamth | oh | 11:16 |
| odyssey4me | winggundamth see https://github.com/openstack/openstack-ansible-os_horizon/commit/452e4e23269dd97e484d4ae1d7050ddb52bc007b | 11:16 |
| odyssey4me | the post install task executes for lbaasv2 which links the panels from an extra installed package to horizon | 11:17 |
| winggundamth | I can see this file in horizon container /openstack/venvs/horizon-13.1.2/lib/python2.7/site-packages/neutron_lbaas_dashboard/enabled/_1481_project_ng_loadbalancersv2_panel.py | 11:18 |
| evrardjp | winggundamth: are you sure you have neutron_lbaas.services.loadbalancer.plugin.LoadBalancerPluginv2 in your neutron_plugin_base AND you don't have something in your user variables that overrides the horizon_enable_neutron_lbaas? | 11:18 |
| winggundamth | evrardjp: yes. I'm sure | 11:18 |
| evrardjp | override to disable | 11:18 |
| evrardjp | ok | 11:18 |
| winggundamth | do you know what is the direct url to access to load balancer page in horizon? | 11:19 |
| evrardjp | could you grep -R horizon_enable_neutron_lbaas * in your config folder ? | 11:19 |
| winggundamth | evrardjp: no I don't see any | 11:21 |
| evrardjp | good | 11:21 |
| evrardjp | did you rerun os-horizon-install.yml ? | 11:21 |
| winggundamth | not yet | 11:21 |
| evrardjp | that's it :D | 11:22 |
| winggundamth | let me rerun it again | 11:22 |
| *** iceyao has quit IRC | 11:23 | |
| winggundamth | evrardjp: done but still no luck :( | 11:24 |
| *** iceyao has joined #openstack-ansible | 11:24 | |
| *** markvoelker has joined #openstack-ansible | 11:25 | |
| winggundamth | I'm sorry but I have to go now. will come back again in next few hours. | 11:25 |
| evrardjp | ok | 11:25 |
| Bofu2U | evrardjp: I removed it from inventory and am re-running setup-everything just so I can get a full picture | 11:25 |
| Bofu2U | fingers crossed :| | 11:25 |
| Bofu2U | I appreciate all of your help, by the way. Thank you. | 11:26 |
| evrardjp | Bofu2U: I hope you have a clean inventory now :D | 11:26 |
| evrardjp | I didn't do anything :p | 11:26 |
| *** shausy has quit IRC | 11:26 | |
| evrardjp | winggundamth: odyssey4me's comment is valuable if you're running mitaka instead of master. in that case, please set horizon_enable_neutron_lbaas: True in your user_variables | 11:27 |
| winggundamth | yes. I'm running mitaka :) | 11:28 |
| winggundamth | 13.1.2 | 11:28 |
| odyssey4me | evrardjp he shouldn't have to - unless there's a logic error | 11:28 |
| odyssey4me | the group_vars should set it automatically | 11:28 |
| odyssey4me | winggundamth can you confirm whether the link is in place to that file you found? | 11:29 |
| winggundamth | odyssey4me: what is the link you refer to? | 11:29 |
| evrardjp | indeed | 11:29 |
| *** markvoelker has quit IRC | 11:29 | |
| evrardjp | even in stable/mitaka | 11:29 |
| winggundamth | root@aio1-horizon-container-44828307:~# ls -l /openstack/venvs/horizon-13.1.2/lib/python2.7/site-packages/neutron_lbaas_dashboard/enabled/ | 11:29 |
| winggundamth | total 8 | 11:29 |
| winggundamth | -rw-r--r-- 1 root root 998 Jun 7 16:25 _1480_project_loadbalancersv2_panel.py | 11:29 |
| winggundamth | -rw-r--r-- 1 root root 1212 Jun 7 16:25 _1481_project_ng_loadbalancersv2_panel.py | 11:29 |
| winggundamth | -rw-r--r-- 1 root root 0 Jun 7 16:25 __init__.py | 11:29 |
| winggundamth | did you mean this? | 11:30 |
| winggundamth | ah need to go now. brb | 11:30 |
| winggundamth | you can leave answer and question here. I'll follow | 11:30 |
| evrardjp | /openstack/venvs/horizon-13.1.2/lib/python2.7/site-packages/openstack_dashboard/local/enabled/ should list the loadbalancersv2 | 11:31 |
| evrardjp | iirc | 11:32 |
| pjm6 | anyone here tried to use novnc of Nova with other keyboard map? | 11:33 |
| evrardjp | yup, good luck | 11:34 |
| evrardjp | :p | 11:34 |
| *** shanemcd has left #openstack-ansible | 11:34 | |
| evrardjp | not supported :p | 11:34 |
| pjm6 | loooool | 11:34 |
| pjm6 | thanks evrardjp :d | 11:34 |
| evrardjp | yw :D | 11:34 |
| pjm6 | i tried changing keymap and other things | 11:34 |
| pjm6 | and didn't worked out | 11:34 |
| evrardjp | lots of quirks | 11:34 |
| pjm6 | how its missing a feature like this? :o | 11:34 |
| evrardjp | consoles are a pain | 11:35 |
| evrardjp | because of ssh? | 11:35 |
| pjm6 | no, no if loose SSH access | 11:35 |
| pjm6 | xD | 11:35 |
| pjm6 | http://docs.openstack.org/admin-guide/compute-remote-console-access.html | 11:36 |
| pjm6 | so this | 11:36 |
| pjm6 | keymap in spice | 11:36 |
| pjm6 | is not working | 11:36 |
| pjm6 | well at least I don't be to bad, because i tried a lot of stuffs and nothing work xD | 11:38 |
| *** sacharya has joined #openstack-ansible | 11:41 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Add metadata checksum fix for AIO-type networks configs https://review.openstack.org/326396 | 11:41 |
| *** LanceHaig has quit IRC | 11:42 | |
| *** LanceHaig has joined #openstack-ansible | 11:42 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_keystone: Minimum example playbook could let suppose db creation https://review.openstack.org/326398 | 11:43 |
| *** sacharya has quit IRC | 11:46 | |
| *** thorst_ has joined #openstack-ansible | 11:46 | |
| Bofu2U | evrardjp: does it matter if I destroy the container, then do the inventory remove or vice versa? | 11:54 |
| Bofu2U | I'm assuming removing it won't destroy the container natively | 11:54 |
| odyssey4me | Bofu2U you can't act on something not in your inventory | 11:54 |
| evrardjp | if you remove it from the inventory you'll have to do it manually | 11:54 |
| odyssey4me | so you should destroy the container first, then remove it from the inventory | 11:54 |
| evrardjp | yeah like odyssey4me said | 11:54 |
| evrardjp | except he types faster | 11:54 |
| Bofu2U | rgr, here we go... | 11:54 |
| evrardjp | but the issue you had was the inventory | 11:55 |
| evrardjp | so you have to clean up as much as you can | 11:55 |
| evrardjp | manually if you need to :p | 11:55 |
| Bofu2U | oh I know, I made a skeleton inventory in case I ever need to restart from galera-only again | 11:56 |
| evrardjp | not sure if that's a right idea | 11:56 |
| *** spotz is now known as spotz_zzz | 11:57 | |
| Bofu2U | I had no choice at the time :-/ | 11:57 |
| evrardjp | if you have a good inventory, it should be fine to target what you want | 11:57 |
| *** psilvad has joined #openstack-ansible | 11:57 | |
| Bofu2U | let me rephrase... I had no inventory at the time | 11:57 |
| Bofu2U | :P | 11:57 |
| *** retreved has joined #openstack-ansible | 11:57 | |
| evrardjp | :D | 11:58 |
| Bofu2U | I'm recovering this cluster from a single backup of a single galera container | 11:58 |
| Bofu2U | and no inventory | 11:58 |
| Bofu2U | lol | 11:58 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Add metadata checksum fix for AIO-type networks configs https://review.openstack.org/326396 | 12:00 |
| odyssey4me | Bofu2U heh, well I guess you're learning a thing or two about disaster recovery ;) | 12:02 |
| Bofu2U | odyssey4me and openstack. Don't forget openstack. | 12:03 |
| *** psilvad has quit IRC | 12:03 | |
| Bofu2U | watching the logs from the neutron agents alone has showed me a bit about how they work | 12:04 |
| Bofu2U | Like ... beyond "oh look networks." | 12:04 |
| odyssey4me | heh, yeah - they're a chatty bunch | 12:04 |
| *** psilvad has joined #openstack-ansible | 12:05 | |
| Bofu2U | it does make me think, though. For the "public" networks, does it still push everything through the controllers even if it's raw VLANs? | 12:05 |
| *** markvoelker has joined #openstack-ansible | 12:06 | |
| odyssey4me | Bofu2U the only thing that goes through the controllers is L3 stuff where the router is controller-managed | 12:06 |
| Bofu2U | So, for example, binding an instance directly onto the public will go direct | 12:06 |
| odyssey4me | so if you have a vlan network setup for a project, and the next hop for the network is set to be something that neutron isn't managing, then nothing will go through the neutron-agent containers | 12:06 |
| Bofu2U | whereas, private network + floating IP will flow through the controller? | 12:06 |
| odyssey4me | yes, private network + floating IP will go through controller because the floating IP NAT rule is implemented on the controller | 12:07 |
| Bofu2U | got it! | 12:08 |
| Bofu2U | so scaling controllers isn't needed as heavily if they're thrown onto public directly | 12:08 |
| odyssey4me | if you're allowing instances to bind to a public network of some sort, and the router for that network is not managed by the controller, then it doesn't go through the l3 agent | 12:08 |
| odyssey4me | well, it may help to know where things are implemented | 12:08 |
| Bofu2U | yeah, obviously | 12:08 |
| odyssey4me | security groups are implemented at the port, in other words on the vif on the hypervisor (compute node) | 12:09 |
| Bofu2U | Now I'm thinking about ... instance put directly onto the public network, but then also added a secondary network interface to private network | 12:09 |
| Bofu2U | so it'd technically ... do ... both? | 12:09 |
| odyssey4me | The floating IP is a NAT rule implemented on the L3 Agent. | 12:09 |
| *** psilvad has quit IRC | 12:10 | |
| odyssey4me | Neutron routers are virtual, and implemented on the L3 Agent. | 12:10 |
| *** zerda2 has quit IRC | 12:10 | |
| odyssey4me | You could possibly have an instance with a leg in a private network and another leg in the public network, but if you have ARP spoofing protection enabled then that instance will not be able to route. It can communicate both ways, but never route. | 12:11 |
| odyssey4me | So that's a useful implementation for something like a proxy (forward or reverse). | 12:11 |
| Bofu2U | I got ya. | 12:11 |
| Bofu2U | Just considering if controllers were to have a problem but some are publicly bound, etc. | 12:12 |
| *** jiteka has joined #openstack-ansible | 12:12 | |
| odyssey4me | Note that the 'public' network is designed for the floating ip pool and virtual routers. If you want to attach instances to a 'public' network then the right thing to implement is a 'shared' network. If you use the 'shared' network you get the same effect and don't have to mangle the policy. | 12:13 |
| Bofu2U | oh that's even better, thank you | 12:13 |
| odyssey4me | Doing that is also a nice way to segregate your problem domains. You know that the 'public' network IP's are all behind a NAT, whereas the 'shared' network IP's are all directly exposed instance NIC's. You may wish to deal with their security profiles differently. | 12:14 |
| Bofu2U | yeah that's perfect | 12:14 |
| Bofu2U | so the real question is - do you want me to take the above and put it into a FAQ Q/A format for you to post somewhere? lol | 12:15 |
| Bofu2U | because I remember trying to find answers like that for a bit | 12:15 |
| *** admin0 has quit IRC | 12:16 | |
| odyssey4me | I wonder if something to that effect shouldn't be added to the networking guide. Maybe here: http://docs.openstack.org/mitaka/networking-guide/intro-os-networking-overview.html | 12:17 |
| Bofu2U | that has my vote .. | 12:18 |
| Bofu2U | just the fact of knowing that the public direct bound doesn't flow through the controller is huge to me | 12:18 |
| Bofu2U | because for some people that could be the difference between 3 controllers and 50 | 12:18 |
| *** wadeholler has quit IRC | 12:19 | |
| Bofu2U | especially since RE: that train of thought, my head jumped to "but if it doesn't go through the controller how does it handle firewall policies" in which you brought up that it's done through the vif on the compute node. | 12:20 |
| odyssey4me | It's kinda described in http://docs.openstack.org/mitaka/networking-guide/scenario-provider-lb.html but I suppose it's a little complex there and a summarised view of the concepts of the 'public' and 'shared' networks would be useful in the overview. | 12:20 |
| odyssey4me | Bofu2U yeah, keep an eye on the terminology there... security groups are not really firewall policies - they're really just access lists | 12:21 |
| Bofu2U | rgr | 12:21 |
| odyssey4me | firewall policies are implemented by FWaaS which is another agent, which would also run on the controller | 12:22 |
| Bofu2U | yeah | 12:22 |
| Bofu2U | example - this is what spawned all of this for me | 12:23 |
| Bofu2U | Right now, everything publicly bound is working 100%. | 12:23 |
| Bofu2U | Since the neutron agent is having issues, everything on a private network and/or floating IP is dead in the water | 12:23 |
| Bofu2U | because the controller 3 agent was set as the "gateway" on the private network I use. That's the one with issues. That's what I'm trying to fix. | 12:24 |
| Bofu2U | etc. | 12:24 |
| *** saneax is now known as saneax_AFK | 12:24 | |
| *** smatzek has quit IRC | 12:26 | |
| *** thorst_ is now known as thorst | 12:31 | |
| *** kylek3h has joined #openstack-ansible | 12:32 | |
| odyssey4me | yeah, as always it really depends on the use case and the equipment available | 12:34 |
| *** saneax_AFK is now known as saneax | 12:35 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create: Make the LXC container create use the host resolver config https://review.openstack.org/326430 | 12:36 |
| openstackgerrit | Greg Faust proposed openstack/openstack-ansible: DOC add note about building AIO more than once https://review.openstack.org/325498 | 12:39 |
| *** admin0 has joined #openstack-ansible | 12:43 | |
| mhayden | morning | 12:45 |
| *** iceyao has quit IRC | 12:45 | |
| mhayden | winggundamth: having problems with LBaaSv2 + Horizon? | 12:45 |
| *** smatzek has joined #openstack-ansible | 12:46 | |
| *** kylek3h has quit IRC | 12:47 | |
| *** wadeholler has joined #openstack-ansible | 12:49 | |
| *** smatzek_ has joined #openstack-ansible | 12:50 | |
| *** smatzek has quit IRC | 12:54 | |
| Bofu2U | mhayden I think he said he'll be back in a few hrs | 12:56 |
| *** kylek3h has joined #openstack-ansible | 12:57 | |
| *** smatzek_ has quit IRC | 12:58 | |
| Adri2000 | hi | 13:00 |
| Bofu2U | hi | 13:02 |
| mhayden | Bofu2U: cool, thanks | 13:04 |
| *** alikins has quit IRC | 13:04 | |
| *** messy has joined #openstack-ansible | 13:05 | |
| Bofu2U | mhayden though once I fix this that's where I'll be next :P | 13:05 |
| *** admin0 has quit IRC | 13:06 | |
| *** alikins has joined #openstack-ansible | 13:07 | |
| *** admin0 has joined #openstack-ansible | 13:12 | |
| openstackgerrit | Greg Faust proposed openstack/openstack-ansible: DOC add note about building AIO more than once https://review.openstack.org/325498 | 13:13 |
| *** Zucan has joined #openstack-ansible | 13:17 | |
| Bofu2U | odyssey4me: what would be the best way to re-add a .1 gateway within a private subnet? | 13:17 |
| Bofu2U | aka it had .1 originally, then I deleted it. Because I didn't want my day to be easy. | 13:18 |
| evrardjp | :/ | 13:18 |
| Bofu2U | (I was trying to change it to a different controller tbh since 3 was screwed, didn't work out so well) | 13:18 |
| evrardjp | At some point don't you want to make your life easy and restart from scratch ? | 13:18 |
| evrardjp | :p | 13:18 |
| *** alikins has quit IRC | 13:18 | |
| Bofu2U | evrardjp nah we're almost back to normal here ;) | 13:18 |
| *** alikins has joined #openstack-ansible | 13:19 | |
| evrardjp | so what did you change ? your external lb vip variable? | 13:19 |
| evrardjp | external/internal? | 13:19 |
| Bofu2U | no, that's all fine | 13:19 |
| evrardjp | or just net configuration on your containers? | 13:19 |
| Bofu2U | I created a network within horizon called "internal" | 13:19 |
| Bofu2U | it made the .1 gateway within neutron, etc. | 13:19 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Docs: Add developer guide for security role https://review.openstack.org/325499 | 13:20 |
| mhayden | asettle: that should address your comments ^^ | 13:20 |
| evrardjp | ok Bofu2U | 13:20 |
| evrardjp | so it's openstack usage now | 13:20 |
| asettle | mhayden: cool :) | 13:20 |
| asettle | And morning | 13:20 |
| Bofu2U | evrardjp: yes, for once :P | 13:20 |
| mhayden | HOWDY ASETTLE | 13:20 |
| automagically | Morning all | 13:20 |
| asettle | yo yo automagically | 13:20 |
| automagically | o/ asettle | 13:20 |
| Bofu2U | mornin automagically asettle | 13:20 |
| automagically | o/ Bofu2U | 13:21 |
| * mhayden passes automagically a breakfast taco | 13:21 | |
| asettle | Also, automagically the other day I kept trying to say "automatically" and all I kept fucking saying was 'automagically' - you have RUINED me | 13:21 |
| asettle | Morning Bofu2U | 13:21 |
| automagically | That’s me, just making a mess everywhere ;) | 13:21 |
| Bofu2U | automagically I'll be saying that after I eat a breakfast taco. | 13:21 |
| asettle | Super frustrating when apparently my job is 'words' | 13:21 |
| asettle | :p | 13:21 |
| mhayden | one must finish consuming breakfast tacos before code can begin merging | 13:21 |
| Bofu2U | or it gets the taco again | 13:22 |
| evrardjp | morning automagically | 13:22 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Add check/audit to gate testing https://review.openstack.org/324482 | 13:22 |
| asettle | Bofu2U: please excuse my ignorance, but I think you're the only person who I don't know? | 13:23 |
| automagically | o/ evrardjp | 13:23 |
| asettle | And your nick isn't giving away any secrets | 13:23 |
| evrardjp | asettle: you mean "word" ? | 13:23 |
| Bofu2U | asettle I'm an unrelated pain in the ass that shows up after creating disaster within my openstack environment hoping to be saved by you kind people. | 13:23 |
| asettle | evrardjp: too right. I'm well gangsta | 13:23 |
| Bofu2U | basically. | 13:23 |
| asettle | Bofu2U: cool. In my head I will call you... Brian | 13:24 |
| Bofu2U | I've also been known to randomly order a crapton of pizza to the rackspace office and have them yell "cloudnull" until he picks it up. | 13:24 |
| Bofu2U | or so I've heard. | 13:24 |
| asettle | hehehehe kevin | 13:25 |
| asettle | But this is giving me no hints | 13:25 |
| asettle | Until then, you're Brian. | 13:25 |
| asettle | Or brain, if i'm dyslexic that day | 13:26 |
| asettle | We'll find someone to be Pinky | 13:26 |
| *** sdake has joined #openstack-ansible | 13:27 | |
| Bofu2U | I'm fine with Brian. | 13:28 |
| Bofu2U | Also, you don't know me. | 13:28 |
| Bofu2U | ;) | 13:28 |
| asettle | ... that's how we got here. | 13:28 |
| Bofu2U | ... indeed. | 13:30 |
| *** maurosr has quit IRC | 13:30 | |
| * Bofu2U affirmative head nod before walking away slowly. | 13:30 | |
| evrardjp | Brian is in the kitchen | 13:30 |
| asettle | :p | 13:30 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Add release note for V-38524 implementation https://review.openstack.org/326467 | 13:31 |
| *** sdake_ has joined #openstack-ansible | 13:34 | |
| mhayden | so i have a patch in to fix one thing that breaks the security role on RHEL 7 -- i can lend someone a RHEL 7 box if they need one to review this patch -> https://review.openstack.org/325421 | 13:35 |
| *** sdake has quit IRC | 13:36 | |
| *** admin0 has quit IRC | 13:38 | |
| *** maurosr has joined #openstack-ansible | 13:40 | |
| *** michaelgugino has joined #openstack-ansible | 13:41 | |
| michaelgugino | morning everyone | 13:41 |
| automagically | o/ michaelgugino | 13:41 |
| *** iceyao has joined #openstack-ansible | 13:41 | |
| asettle | o/ | 13:42 |
| *** admin0 has joined #openstack-ansible | 13:43 | |
| automagically | I’d appreciate some core reviews on https://review.openstack.org/325426 please if you all have a moment | 13:43 |
| automagically | Also, https://review.openstack.org/325419 | 13:44 |
| automagically | Thanks in advance | 13:44 |
| *** admin0 has quit IRC | 13:45 | |
| *** KLevenstein has joined #openstack-ansible | 13:46 | |
| *** admin0 has joined #openstack-ansible | 13:50 | |
| mhayden | automagically: i shall gandereth | 13:50 |
| *** ametts has joined #openstack-ansible | 13:53 | |
| *** smatzek_ has joined #openstack-ansible | 13:59 | |
| *** spotz_zzz is now known as spotz | 13:59 | |
| odyssey4me | Bofu2U have you figured out your .1 gateway thingy yet? | 14:00 |
| *** pester has quit IRC | 14:00 | |
| odyssey4me | Bofu2U if not, what's the actual situation - you had a virtual network with a virtual router (which had the .1 address), and now have a vlan and want to set it up the same way except have .1 be on a real router? | 14:01 |
| *** KLevenstein has quit IRC | 14:02 | |
| *** KLevenstein has joined #openstack-ansible | 14:04 | |
| *** Mudpuppy has joined #openstack-ansible | 14:04 | |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts: Fail immediately when kernel needs an update https://review.openstack.org/325419 | 14:04 |
| *** Mudpuppy has quit IRC | 14:04 | |
| *** Mudpuppy has joined #openstack-ansible | 14:05 | |
| Bofu2U | odyssey4me: virtual network with virtual router and just need to setup the virtual router again on .1 | 14:07 |
| Bofu2U | that's all | 14:07 |
| odyssey4me | Bofu2U ok, if you do a port list on that network - do any of the ports already have .1 assigned to it? | 14:08 |
| Bofu2U | nope | 14:08 |
| odyssey4me | ok, and do you already have a router with an internal interface on that network? | 14:08 |
| odyssey4me | (and it has the wrong address) | 14:08 |
| Bofu2U | not yet | 14:08 |
| Bofu2U | I didn't want to potentially make it worse by messing with randomness | 14:08 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-galera_server: DOC - Adopting the common role documentation pattern https://review.openstack.org/326492 | 14:09 |
| odyssey4me | alright, if you create a new router it should give you the option of creating it with a specific address on that network | 14:09 |
| odyssey4me | you can certainly do so when creating a port | 14:09 |
| Bofu2U | k let me log back in and give it a shot | 14:10 |
| *** jamesdenton has joined #openstack-ansible | 14:11 | |
| Bofu2U | as my neutron agent keeps spamming about agents offline and migrations :| | 14:11 |
| mrhillsman | morning | 14:11 |
| spotz | morning | 14:11 |
| Bofu2U | morning | 14:11 |
| odyssey4me | asettle can you cast your beady eye over https://review.openstack.org/325498 quickly please? | 14:14 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_keystone: Minimum example playbook could let suppose db creation https://review.openstack.org/326398 | 14:17 |
| *** admin0 has quit IRC | 14:18 | |
| *** psilvad has joined #openstack-ansible | 14:19 | |
| Bofu2U | odyssey4me I added it back in... says status down... hm | 14:19 |
| Bofu2U | admin state up, status down | 14:20 |
| mhayden | automagically: sorry, running behind on those reviews i promised | 14:21 |
| Bofu2U | odyssey4me: to clarify, on the port after adding the interface in | 14:22 |
| *** mpjetta has joined #openstack-ansible | 14:22 | |
| openstackgerrit | Wang Qing wu proposed openstack/openstack-ansible-os_nova: Implement Nova PowerVM Virt Driver https://review.openstack.org/319022 | 14:23 |
| odyssey4me | Bofu2U depending on how many routers and networks you have, the agent may take a bit of time before it gets to create the port properly - otherwise check the l3 agent logs for any errors to see why its down | 14:24 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-openstack_hosts: Updated the hostname generation https://review.openstack.org/323504 | 14:24 |
| Bofu2U | not even seeing the l3 agent running in the container o.O | 14:24 |
| Bofu2U | well wait it is... but no logs | 14:25 |
| *** klamath has quit IRC | 14:25 | |
| *** klamath has joined #openstack-ansible | 14:25 | |
| odyssey4me | Bofu2U ah, you may need to flip it into debug mode | 14:25 |
| odyssey4me | Bofu2U but you may also find logs somewhere else that give clues - like the kernel log or syslog | 14:25 |
| *** jamesdenton has quit IRC | 14:30 | |
| Bofu2U | odyssey4me hm. that's just filled with debugs about "agent down migrating..." | 14:31 |
| Bofu2U | which makes sense given the status | 14:31 |
| *** sdake_ has quit IRC | 14:36 | |
| *** weezS has joined #openstack-ansible | 14:36 | |
| *** raddaoui has joined #openstack-ansible | 14:39 | |
| *** mfisch has joined #openstack-ansible | 14:39 | |
| *** mfisch has quit IRC | 14:40 | |
| *** mfisch has joined #openstack-ansible | 14:40 | |
| *** galstrom_zzz is now known as galstrom | 14:42 | |
| *** admin0 has joined #openstack-ansible | 14:42 | |
| flaviodsr | odyssey4me can you take a look at http://logs.openstack.org/73/325673/1/check/gate-openstack-ansible-os_sahara-docs/305c8f5/console.html ? Why this is failing? I am totally lost here | 14:43 |
| *** bryan_att has quit IRC | 14:44 | |
| *** admin0 has quit IRC | 14:44 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-lxc_container_create: DOC - Adopting the common role documentation pattern https://review.openstack.org/326520 | 14:45 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Extract and test inventory and backup I/O https://review.openstack.org/323601 | 14:45 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Enable human readable logging https://review.openstack.org/321333 | 14:46 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-openstack_hosts: Cleanup/standardize usage of tags https://review.openstack.org/325418 | 14:48 |
| Bofu2U | evrardjp: hm. looks like "neutron agent-list" doesn't have the right container IDs as the running containers... inventory problem? :P | 14:50 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance: Implement CentOS 7 support in os_glance https://review.openstack.org/320542 | 14:54 |
| *** admin0 has joined #openstack-ansible | 14:54 | |
| *** weshay has quit IRC | 14:56 | |
| *** bryan_att has joined #openstack-ansible | 14:57 | |
| odyssey4me | flaviodsr looking... | 15:00 |
| odyssey4me | Bofu2U if you essentially rebuilt the environment, but kept the old DB, then I expect that you'll still have the old agents around in the DB and will need to disable them to prevent networks/routers to be scheduled to use them | 15:01 |
| Bofu2U | odyssey4me right that's what I'm doing - but the new ones aren't re-adding themselves either | 15:01 |
| Nepoc | FWIW whoever saw my last message regarding neutron acting unusual after the install... I've had a to add in a restart of all neutron services in all containers and compute nodes to fix everything. Why I have to do this is still a mystery. | 15:04 |
| asettle | Sorry odyssey4me I was in a meeting previously. But the patch LGTM. | 15:05 |
| asettle | I'd reviewed it earlier and all iterations had been included :) | 15:05 |
| asettle | Give or take one or two | 15:05 |
| asettle | But it's been merged, so that's okay :) | 15:06 |
| *** vnogin has joined #openstack-ansible | 15:10 | |
| *** saneax is now known as saneax_AFK | 15:10 | |
| *** Drago has joined #openstack-ansible | 15:10 | |
| *** Drago has quit IRC | 15:12 | |
| *** vnogin1 has quit IRC | 15:12 | |
| *** Drago has joined #openstack-ansible | 15:12 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-lxc_hosts: DOC - Adopting the common role documentation pattern https://review.openstack.org/326568 | 15:13 |
| Bofu2U | odyssey4me so this is interesting... my instances and private networks just came up, but neutron agent-list shows xxx on all instances lol | 15:17 |
| odyssey4me | flaviodsr OK, the issue is that ansible relies on paramiko and the latest version of paramiko requires some libraries to be present - so ensure that https://github.com/openstack/openstack-ansible-os_sahara/blob/master/other-requirements.txt includes the extra content that https://github.com/openstack/openstack-ansible-galera_server/blob/master/other-requirements.txt has | 15:17 |
| Bofu2U | odyssey4me scratch that ... apparently at some point things switched to underscores instead of dashes so my grep was screwed | 15:18 |
| odyssey4me | flaviodsr that will ensure that zuul installs those packages onto the node before executing the test job | 15:18 |
| odyssey4me | Bofu2U hmmm, FYI we switched FROM underscores TO dashes for Mitaka... but the 13.0.0 release still used underscores | 15:19 |
| odyssey4me | 13.0.1 included the switch | 15:19 |
| odyssey4me | Bofu2U https://review.openstack.org/323504 is a patch to make them all aliases so that we can all get along | 15:19 |
| *** cloudtrainme has joined #openstack-ansible | 15:22 | |
| Bofu2U | odyssey4me gracias | 15:23 |
| flaviodsr | odyssey4me got it, thanks! | 15:24 |
| *** deadnull is now known as deadnull_ | 15:25 | |
| *** weshay has joined #openstack-ansible | 15:27 | |
| Bofu2U | odyssey4me evrardjp lbragstad I'm back up and running, thank you all very much. | 15:28 |
| evrardjp | good news | 15:28 |
| Bofu2U | I should be back in a few days after killing it again. | 15:28 |
| Bofu2U | :P | 15:28 |
| *** sdake has joined #openstack-ansible | 15:28 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible: fix links in upgrade-playbooks https://review.openstack.org/326580 | 15:31 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone: Add support for CustomLog format modification https://review.openstack.org/325837 | 15:31 |
| *** mikelk has quit IRC | 15:33 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone: Remove pip_lock_down dependency https://review.openstack.org/313889 | 15:34 |
| *** chhavi has quit IRC | 15:38 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone: Minimum example playbook could let suppose db creation https://review.openstack.org/326398 | 15:38 |
| odyssey4me | jmccrory automagically cloudnull d34dh0r53 stevelle mattt hughsaunders andymccr mhayden any thoughts on https://review.openstack.org/326396 ? also, https://review.openstack.org/326430 needs a review or two please | 15:39 |
| * automagically looking | 15:40 | |
| *** admin0 has quit IRC | 15:40 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron: Updated multi-distro setup for isolation https://review.openstack.org/324974 | 15:41 |
| *** sacharya has joined #openstack-ansible | 15:42 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server: DOC - Adopting the common role documentation pattern https://review.openstack.org/326492 | 15:46 |
| *** Kdecherf has left #openstack-ansible | 15:47 | |
| *** sacharya has quit IRC | 15:47 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova: Fix nova_system_group_uid variable name https://review.openstack.org/326125 | 15:48 |
| *** TheIntern has joined #openstack-ansible | 15:48 | |
| *** johnmilton has quit IRC | 15:49 | |
| *** johnmilton has joined #openstack-ansible | 15:50 | |
| *** BjoernT has joined #openstack-ansible | 15:51 | |
| *** smatzek_ has quit IRC | 15:52 | |
| *** johnmilton has quit IRC | 15:56 | |
| *** TheIntern has quit IRC | 15:56 | |
| odyssey4me | bug triage here cloudnull, mattt, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, Sam-I-Am, odyssey4me, serverascode, rromans, erikmwilson, mancdaz, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, mhayden, scarlisle, luckyinva, ntt, javeriak, automagically, spotz, vdo, jmccrory, alextricity25, jasondotstar, KLevenstein, admin0, michaelgugino, ametts, v1k0d3n, | 16:00 |
| odyssey4me | severion, bgmccollum, darrenc, JRobinson__, asettle, colinmcnamara, thorst, adreznec | 16:00 |
| michaelgugino | here | 16:00 |
| spotz | \o/ | 16:00 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Add metadata checksum fix for AIO-type networks configs https://review.openstack.org/326396 | 16:00 |
| thorst | \o | 16:00 |
| adreznec | o/ | 16:00 |
| v1k0d3n | here, but have to step away... | 16:00 |
| evrardjp | o/ | 16:01 |
| odyssey4me | before we start, I'd like to ask if there are any volunteers who'd be happy to facilitate the bug triage on a weekly basis | 16:01 |
| evrardjp | I'd be happy to help OSA, if I can skip this week :p | 16:01 |
| prometheanfire | neato | 16:01 |
| odyssey4me | ideally I'd like it to rotate, but I'd like to see someone take a primary triage role and to ensure that there are a facilitator for every meeting | 16:01 |
| automagically | o/ | 16:02 |
| asettle | o/ | 16:02 |
| evrardjp | if you're alright with that odyssey4me | 16:03 |
| odyssey4me | evrardjp Sure, that'd be great! | 16:03 |
| *** sacharya has joined #openstack-ansible | 16:03 | |
| odyssey4me | It'd help me greatly. It might be a nice opportunity to try and document the meeting's guidelines a little more in the contributor guide so that any facilitators and contributors are aware of what happens and how it all works. | 16:04 |
| automagically | Thanks for volunteering evrardjp | 16:04 |
| openstackgerrit | Steve Lewis (stevelle) proposed openstack/openstack-ansible-os_gnocchi: Implement Ubuntu 16.04 support with SystemD https://review.openstack.org/324715 | 16:04 |
| evrardjp | yw | 16:04 |
| odyssey4me | righto, on to the list | 16:05 |
| odyssey4me | https://bugs.launchpad.net/openstack-ansible/+bugs?search=Search&field.status=New | 16:05 |
| odyssey4me | I'll start at the bottom and work up. | 16:05 |
| odyssey4me | https://bugs.launchpad.net/openstack-ansible/+bug/1588902 | 16:05 |
| openstack | Launchpad bug 1588902 in openstack-ansible "Variables defined in global_overrides persist after deletion" [Undecided,New] | 16:05 |
| automagically | I consider that a high/medium given the confusion/dismay/despair it can cause a deployer | 16:06 |
| automagically | :D | 16:06 |
| odyssey4me | heh, oh dear - editing any vars in openstack_user_config will result in the same behaviour due to the fact that openstack_user_config forms part of the inventory | 16:06 |
| odyssey4me | these vars should be set in user_variables instead | 16:06 |
| automagically | Yep | 16:06 |
| odyssey4me | I suppose that we're not clear about this. | 16:07 |
| gregfaust | i have been bitten by 1588902 when testing various proxy & mirror settings | 16:07 |
| odyssey4me | palendae ping? | 16:07 |
| palendae | Pong | 16:08 |
| odyssey4me | So I can see that for some hosts you may want to set that var on a per host basis, depending on your environment. | 16:08 |
| palendae | automagically, Are you taking that? | 16:08 |
| palendae | Or did you decided not to? | 16:08 |
| *** johnmilton has joined #openstack-ansible | 16:08 | |
| automagically | palendae: I haven’t gotten to it as of yet | 16:08 |
| automagically | So, its up for grabs | 16:08 |
| palendae | ok | 16:08 |
| odyssey4me | So while a 'standard' environment would have it in user_variables, it seems that we should probably figure out how to deal with the persistance in the inventory too? | 16:08 |
| *** deadnull_ is now known as deadnull | 16:08 | |
| odyssey4me | palendae one of the items we described in the 'wish list' for the inventory was awareness of a change of state | 16:09 |
| palendae | Well this is clearly a bug | 16:09 |
| palendae | Dunno if it's even a wish list | 16:09 |
| odyssey4me | ie you have this setting now, then you remove it or change it - especially in the inventory this is important | 16:09 |
| stevelle | lots of ways to skin this one | 16:09 |
| *** smatzek_ has joined #openstack-ansible | 16:09 | |
| palendae | This could apply to changing networks too | 16:09 |
| odyssey4me | do you think we can try to build some sort of state change awareness into the inventory? | 16:10 |
| evrardjp | Confirm/Medium? | 16:10 |
| palendae | Which is probably the more common case for global_overrides | 16:10 |
| odyssey4me | yep, changing networks is another state change which would be important - perhaps virt_type changes too | 16:10 |
| palendae | Would have to look - right now all it's doing is re-reading the .json file | 16:10 |
| palendae | So if the inventory.json file exists, it wins by virtue of nothing else being read | 16:10 |
| *** smatzek_ has quit IRC | 16:10 | |
| odyssey4me | ok - so do we have two items from this? the one would be that we should document that this should be done in user_variables.yml as a standard way of implementing it | 16:11 |
| palendae | Since the .yml files aren't touched, this may be a fairly large fix, but we definitely need to address it | 16:11 |
| palendae | odyssey4me, What is "this" that should be done? | 16:11 |
| odyssey4me | the other would be to detect the removal of a var from the inventory and to remove it from the inventory.json if that happens? | 16:11 |
| palendae | This affects stuff that goes in global_overrides but not user_variables, like networks | 16:11 |
| automagically | Definitely a doc update or two needed here | 16:12 |
| palendae | I can take your latter case | 16:12 |
| odyssey4me | palendae by 'this' I mean the use of the var in the inventory/openstack_user_config | 16:12 |
| palendae | Which var? | 16:12 |
| palendae | This particular bug report shows a strawman | 16:12 |
| palendae | automagically, What was the actual var you were using? | 16:12 |
| odyssey4me | I expect that it was https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/user_variables.yml#L137-L146 | 16:13 |
| automagically | I don’t recall to be honest, its on the other side of the weekend | 16:13 |
| automagically | odyssey4me: Yeah, that looks likely | 16:13 |
| palendae | Ok | 16:13 |
| palendae | IMO we'd be best off advising nothing goes in global_overrides except networks | 16:13 |
| evrardjp | palendae: agreed | 16:14 |
| odyssey4me | ok, we should probably then register a doc bug for that specifically | 16:14 |
| evrardjp | if it's global it can be defined in user_* | 16:14 |
| palendae | So yes, I would agree it's 2 bugs - doc and the regeneration issue | 16:14 |
| palendae | And I can snag the regeneration part | 16:14 |
| odyssey4me | can someone please register a bug for the recommended doc change? | 16:14 |
| spotz | odyssey4me: I'll go through the log and do it | 16:15 |
| odyssey4me | thanks, I've added a comment to https://bugs.launchpad.net/openstack-ansible/+bug/1588902 for clarity | 16:17 |
| openstack | Launchpad bug 1588902 in openstack-ansible "Variables defined in global_overrides persist after deletion" [Undecided,New] | 16:17 |
| odyssey4me | thanks spotz :) | 16:17 |
| odyssey4me | palendae can you self assign please | 16:17 |
| palendae | Sur | 16:17 |
| odyssey4me | thoughts on importance for this one? | 16:17 |
| evrardjp | importance: wishlist | 16:17 |
| palendae | I'd say medium or high | 16:17 |
| palendae | Because this is the opposite of expected behavior, and prevents deployers from managing their environment without getting invasive | 16:18 |
| odyssey4me | yeah ok - we have two suggestions for medium/high... I'm thinking medium as this is a case of not using by design and is a bit of a fringe case | 16:19 |
| odyssey4me | that said, it's a large environment use-case so I'd be happy to go with high | 16:19 |
| *** asettle has quit IRC | 16:19 | |
| odyssey4me | stevelle thoughts? | 16:19 |
| stevelle | I'm feeling like it doesn't matter much either way, but I'd grant this high, because the workarounds are not yet documented well | 16:20 |
| *** karimb has joined #openstack-ansible | 16:20 | |
| odyssey4me | ok, done | 16:21 |
| odyssey4me | moving on: https://bugs.launchpad.net/openstack-ansible/+bug/1588821 | 16:21 |
| openstack | Launchpad bug 1588821 in openstack-ansible "Document host removal" [Undecided,New] | 16:21 |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible: DOC - fix links in upgrade-playbooks https://review.openstack.org/326580 | 16:21 |
| evrardjp | my wishlist was for this one | 16:23 |
| evrardjp | :p | 16:23 |
| evrardjp | high for the previous one :p | 16:24 |
| odyssey4me | it is obviously clear that it would be useful to have the decommission process documented properly | 16:24 |
| odyssey4me | we could probably also add a convenience playbook to the ops repository | 16:24 |
| odyssey4me | any volunteers to pick this up? | 16:25 |
| odyssey4me | ok, marking wishlist and confirmed for now | 16:27 |
| odyssey4me | next up https://bugs.launchpad.net/openstack-ansible/+bug/1588818 | 16:27 |
| openstack | Launchpad bug 1588818 in openstack-ansible "Make network and subnet names configurable in tempest installation" [Undecided,New] | 16:27 |
| odyssey4me | it looks like a low hanging fruit enhancement | 16:28 |
| odyssey4me | so wishlist, clearly | 16:29 |
| odyssey4me | next up: https://bugs.launchpad.net/openstack-ansible/+bug/1588812 | 16:29 |
| openstack | Launchpad bug 1588812 in openstack-ansible "br-vlan should not be required on compute nodes" [Undecided,New] | 16:29 |
| stevelle | that's legitimate, isn't it | 16:32 |
| evrardjp | it legitimate if we use br-vlan on neutron but not on compute | 16:32 |
| odyssey4me | sort-of related: https://review.openstack.org/276320 | 16:32 |
| stevelle | related, but distinct still. | 16:33 |
| odyssey4me | yeah | 16:33 |
| odyssey4me | so the use-case is right - and we do have a rather fixed assumption of expected bridges, etc | 16:34 |
| *** Nepoc has quit IRC | 16:34 | |
| stevelle | I'd got with confirmed low | 16:34 |
| stevelle | s/t// | 16:34 |
| evrardjp | +1 | 16:34 |
| odyssey4me | it would be nice if we could move towards using the more standard bridges described in the network guide, but that's going to take quite a bit of re-work and dedicated time. | 16:34 |
| evrardjp | agreed | 16:36 |
| odyssey4me | ok, marked confirmed/low | 16:36 |
| odyssey4me | next up: https://bugs.launchpad.net/openstack-ansible/+bug/1588777 | 16:36 |
| openstack | Launchpad bug 1588777 in openstack-ansible "Cinder volume type creation race condition failure" [Undecided,New] - Assigned to Samuel Matzek (smatzek) | 16:36 |
| odyssey4me | hrm, so the only way to work around that right now is to make https://github.com/openstack/openstack-ansible-os_cinder/blob/master/tasks/cinder_backends.yml#L27-L37 only run on the first member of the group | 16:38 |
| evrardjp | you should not run this way IMO | 16:38 |
| odyssey4me | or that task should come out of the role and go into the playbook | 16:38 |
| stevelle | so this is during the configuration of cinder, really. | 16:38 |
| *** admin0 has joined #openstack-ansible | 16:42 | |
| odyssey4me | confirmed/medium in my opinion | 16:42 |
| evrardjp | confirmed medium | 16:42 |
| odyssey4me | I've added a comment regarding the options I see to solve it | 16:43 |
| odyssey4me | next up: https://bugs.launchpad.net/openstack-ansible/+bug/1588051 | 16:43 |
| openstack | Launchpad bug 1588051 in openstack-ansible "Excessive logging to /var/log in neutron agents and other containers" [Undecided,New] | 16:43 |
| *** karimb has quit IRC | 16:45 | |
| *** cloudtrainme has quit IRC | 16:45 | |
| *** karimb has joined #openstack-ansible | 16:45 | |
| evrardjp | it's a wishlist for me | 16:46 |
| odyssey4me | well, how a host is partitioned and where the logs go is not really something the project can solve | 16:46 |
| evrardjp | with even a proposition of how to fix it | 16:46 |
| odyssey4me | what we can do, however, is implement facilities that deployers can consume | 16:46 |
| BjoernT | that's a funny opinion | 16:46 |
| odyssey4me | cloudnull has made a reasonable suggestion | 16:47 |
| BjoernT | OSA rolls out a design in regards to logging | 16:47 |
| odyssey4me | any other suggestions? | 16:47 |
| BjoernT | so we can not just look away. Yes as I updated. Do not store any logs in the root partitions | 16:47 |
| *** karimb has quit IRC | 16:48 | |
| BjoernT | Basically what I updated to the beginning of the meeting. | 16:48 |
| evrardjp | BjoernT: is that something you see impacting 100% of the osa customers? | 16:48 |
| odyssey4me | BjoernT where the logs go in terms of partitions depends on how you partition your hosts | 16:48 |
| BjoernT | My customers are | 16:48 |
| odyssey4me | if you partition /openstack/logs then your logs aren't in the root partition. | 16:48 |
| evrardjp | I don't ask names :D | 16:48 |
| *** evilt0ne has quit IRC | 16:48 | |
| BjoernT | yeah a stop gap would be to bind mount /var/log | 16:49 |
| BjoernT | that keeps filling up not so much neutron | 16:49 |
| evrardjp | BjoernT: I think what you're pointing is another issue | 16:49 |
| *** Drago has quit IRC | 16:49 | |
| BjoernT | the issue is that the DHCP lease files are on the container root disks which is a different problem | 16:50 |
| evrardjp | why would a log only be in the container instead of being in /openstack/logs ? | 16:50 |
| evrardjp | right? | 16:50 |
| BjoernT | wrong | 16:50 |
| evrardjp | then we have a bug of a file not correctly reported? | 16:50 |
| odyssey4me | alright, so it seems that we have logs leaking into syslog/upstart logs which is one issue | 16:50 |
| BjoernT | the openstack/neutron logs are at /openstack/logs | 16:50 |
| evrardjp | BjoernT: that's what I meant | 16:50 |
| evrardjp | with the dhcp | 16:50 |
| BjoernT | the DHCP lease files are on /dev/root | 16:50 |
| *** Drago has joined #openstack-ansible | 16:50 | |
| *** Drago has quit IRC | 16:50 | |
| BjoernT | to say it that way | 16:50 |
| *** iceyao has quit IRC | 16:51 | |
| *** Drago has joined #openstack-ansible | 16:51 | |
| *** Drago has quit IRC | 16:51 | |
| BjoernT | and all other rsyslog logs are on /dev/root too which ultimately can fill up disk space uncontrolled like conntrackd | 16:51 |
| *** Drago has joined #openstack-ansible | 16:51 | |
| BjoernT | same is true for neutron pid files, namespaces etc all this stuff is on /dev/root and potentially in danger if conntrackd freaks out | 16:51 |
| BjoernT | hence NO logging on /dev/root | 16:52 |
| odyssey4me | by that you mean the root disk in the container? | 16:52 |
| BjoernT | correct | 16:52 |
| BjoernT | it's easier to type/write | 16:52 |
| BjoernT | and everyone else will understand it :-) | 16:52 |
| odyssey4me | ok - I'm clarifying to ensure that we understand the problem properly | 16:52 |
| BjoernT | I think the issue is compounded using LVM as storage backend | 16:53 |
| stevelle | it had to be clarified so clearly not everyone understood :) | 16:53 |
| odyssey4me | :p | 16:53 |
| * stevelle raises his hand | 16:53 | |
| odyssey4me | go stevelle | 16:53 |
| BjoernT | stevelle: you need a linux course, I can make a training for you, lol | 16:53 |
| *** admin0 has quit IRC | 16:53 | |
| *** admin0 has joined #openstack-ansible | 16:54 | |
| *** psilvad has quit IRC | 16:54 | |
| odyssey4me | ok, so the trouble with moving the logs to a bind mount on the host is that the problem shifts from one container to all of them | 16:54 |
| odyssey4me | if you fill up the host, you kill the whole host | 16:54 |
| BjoernT | yeah so that's where I want to go. Get all content likle logs, config , all volatile data out of the container but this is a bigger picture. | 16:55 |
| odyssey4me | logrotate is a help, and perhaps should allow configuration based on size instead of time | 16:55 |
| BjoernT | odysee4me, we can partition /var/lib/lxc accordingly and have monitoring in place | 16:55 |
| evrardjp | I think it starts to be a discussion based on opinions | 16:55 |
| odyssey4me | but as you say logrotate doesn't cover all the bases | 16:55 |
| *** psilvad has joined #openstack-ansible | 16:55 | |
| *** admin0 has quit IRC | 16:55 | |
| BjoernT | Yes but logrotate is also not guaranteed to succeed | 16:56 |
| openstackgerrit | Merged openstack/openstack-ansible: RFC1034/5 hostname upgrade https://review.openstack.org/323033 | 16:56 |
| openstackgerrit | Merged openstack/openstack-ansible: DOC add note about building AIO more than once https://review.openstack.org/325498 | 16:56 |
| evrardjp | because you're talking about you can solving that with monitoring too, and there is a risk of impacting the whole host. For me it looks interesting, but everything should be done facultatively | 16:56 |
| BjoernT | I also do see double logging like /var/log/upstart and /openstack/log etc | 16:56 |
| openstackgerrit | Merged openstack/openstack-ansible: Update ansible to version 2.1 https://review.openstack.org/321042 | 16:56 |
| openstackgerrit | Merged openstack/openstack-ansible: change host_containers group names in inventory https://review.openstack.org/283149 | 16:56 |
| *** admin0 has joined #openstack-ansible | 16:56 | |
| openstackgerrit | Merged openstack/openstack-ansible: Use combined pip_install role https://review.openstack.org/313878 | 16:56 |
| odyssey4me | another option entirely is to ensure that there's a configuration ability to never log locally and only send logs via rsyslog... but that has other issues | 16:56 |
| BjoernT | evrardjp: Every discussion starts with a opinionated problem , what's wrong with that. I'm as biggest OSA user have clearly a problem here | 16:57 |
| odyssey4me | yeah, so the double-logging and log leakage to syslog is something that should be addressed in the short term | 16:57 |
| odyssey4me | I think that better options for handling log sizes via logrotate is another way to help band aid the issue. | 16:58 |
| BjoernT | odyssey4me: Yes I was curios what we think in not using log files in log4python or whatever this facility is named and directly ship all logs to a syslog facility so I can decided where stuff goes to | 16:58 |
| odyssey4me | BjoernT the main reason we're logging locally is because the openstack logging direct to syslog has been very broken for a long time | 16:58 |
| openstackgerrit | Michael Gugino proposed openstack/openstack-ansible-os_nova: Implement Xenial Support https://review.openstack.org/324045 | 16:59 |
| BjoernT | bugs in the python rsyslog implementation ? | 16:59 |
| odyssey4me | so we opted to log locally, then use rsyslog to pick up the log and ship it instead - it's more reliable, and lets the service run regardless of the remote rsyslog availability | 16:59 |
| odyssey4me | it's a more robust implementation | 16:59 |
| odyssey4me | BjoernT no, bugs in the openstack usage of the underlying libraries | 16:59 |
| odyssey4me | some worked fine, others did not - and we opted for consistency in the implementation rather than doing different things for different services | 17:00 |
| *** weezS has quit IRC | 17:00 | |
| odyssey4me | nova, for instance, if it could not connect to the syslog device, would simply hang when starting | 17:00 |
| odyssey4me | it would never fail, and never give up trying | 17:00 |
| odyssey4me | it's plausible that the situation has changed - but this was still an issue when I last tested in Liberty | 17:01 |
| evrardjp | I like the discussion, but we have to understand that this is a big move, and this requires careful planning. It requires a spec for the analysis, and is (IMO) a wishlist | 17:01 |
| evrardjp | it's not a bug: we have logging in place, and with proper sysadmin behavior, nothing brings a general failure itself | 17:02 |
| evrardjp | still this is my opinion only :) | 17:02 |
| BjoernT | evrardjp: yes as I offered I can start writing up specs once we are in line to a) identify this is a problem. | 17:03 |
| BjoernT | In the past I wasted time discussing things and solutions which was not even accepted as problem, I don't want to get caught into this again it is frustrating | 17:03 |
| evrardjp | makes sense | 17:03 |
| BjoernT | so next thing is i write up a spec ? | 17:05 |
| odyssey4me | marking as confirmed/low priority as there are clear workaround available and I've added a link to this discussion with the suggestion to submit a spec for the larger picture of better log management | 17:05 |
| odyssey4me | BjoernT sure - the spec template includes a decription of the problem | 17:06 |
| BjoernT | that's Ok this is what I offered | 17:06 |
| odyssey4me | so there may be some discussion/iteration on what the problem is | 17:06 |
| odyssey4me | once the problem itself is agreed on and a suitable solution is proposed then it can be worked on | 17:07 |
| BjoernT | yes we can do that in scope of the spec | 17:07 |
| BjoernT | to make it clear | 17:07 |
| odyssey4me | for now though this bug can have some band-aids implemented | 17:07 |
| odyssey4me | as we discussed earlier | 17:07 |
| BjoernT | yes as short gap this is fine | 17:07 |
| *** Drago has quit IRC | 17:08 | |
| odyssey4me | I think regardless of a long term solution, the double-logging needs to be addressed. It wreaks all sort of havoc with any form of log analysis. | 17:09 |
| BjoernT | yes correct | 17:10 |
| odyssey4me | alright, we're out of time for this week - thanks all for attending and participating in the discussion | 17:10 |
| odyssey4me | evrardjp will be leading the triage from next week onwards :) | 17:10 |
| evrardjp | thanks everyone | 17:10 |
| *** zerda2 has joined #openstack-ansible | 17:18 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Set check_mode variable every time https://review.openstack.org/326664 | 17:19 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_aodh: DOC - Adding details on required variables for the role https://review.openstack.org/326665 | 17:20 |
| zerda2 | Hello. Have some questions regarding https://review.openstack.org/#/c/325837/. I'd like to have this change in Liberty, and also in Horizon (since it's basically the same Apache+WSGI thing). Should I create 5 more reviews (for keystone L, M and Horizon L, M, N)? | 17:21 |
| automagically | zerda2: You can propose a cherry-pick using the gerrit web interface to the stable/mitaka and liberty branchs | 17:22 |
| automagically | And I think adding a similar change to horizon is a good idea | 17:23 |
| stevelle | zerda2: ^ cherry-pick for mitaka. For liberty the roles are all in the main openstack-ansible repo | 17:23 |
| stevelle | a manual backport will be required there | 17:23 |
| zerda2 | So, one cherrypick and one manual backport per module | 17:23 |
| stevelle | that sounds right | 17:24 |
| zerda2 | Cool, will try to get to this tomorrow | 17:24 |
| automagically | Awesome zerda2 , thanks for the contrib | 17:25 |
| *** Ashana has joined #openstack-ansible | 17:28 | |
| zerda2 | automagically, it's been a pleasure, I'm really amazed with the reaction speed on this change. We (my employer) are using openstack-ansible for our internal installations, and it's great to have necessary modifications in upstream in timely manner, not some forks and patches maintained locally | 17:28 |
| *** admin0 has quit IRC | 17:30 | |
| *** woodard has joined #openstack-ansible | 17:30 | |
| *** admin0 has joined #openstack-ansible | 17:31 | |
| zerda2 | And speaking about Horizon logging, does it make sense to change it like in python-horizon module for better control (https://github.com/openstack/puppet-horizon/blob/master/templates/local_settings.py.erb#L654)? Now there are messages in Apache error log saying "Login successful for user X". | 17:32 |
| *** cloudtrainme has joined #openstack-ansible | 17:32 | |
| zerda2 | And that's because of console handler hardcoded into template | 17:33 |
| *** admin0 has quit IRC | 17:34 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts: DOC - Adopting the common role documentation pattern https://review.openstack.org/326568 | 17:37 |
| *** admin0 has joined #openstack-ansible | 17:39 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create: Make the LXC container create use the host resolver config https://review.openstack.org/326430 | 17:39 |
| *** woodard has quit IRC | 17:41 | |
| *** admin0 has quit IRC | 17:41 | |
| *** woodard has joined #openstack-ansible | 17:41 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Remove deleted override vars from inventory https://review.openstack.org/325380 | 17:42 |
| *** admin0 has joined #openstack-ansible | 17:42 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Remove deleted override vars from inventory https://review.openstack.org/325380 | 17:45 |
| openstackgerrit | Steve Lewis (stevelle) proposed openstack/openstack-ansible: Introduce a playbook for deploying Gnocchi https://review.openstack.org/324125 | 17:45 |
| *** woodard has quit IRC | 17:45 | |
| palendae | Lunch and relocating | 17:46 |
| *** asettle has joined #openstack-ansible | 17:49 | |
| *** asettle has quit IRC | 17:49 | |
| *** asettle has joined #openstack-ansible | 17:49 | |
| *** admin0 has quit IRC | 17:50 | |
| *** admin0 has joined #openstack-ansible | 17:52 | |
| *** cloudtrainme has quit IRC | 17:53 | |
| *** Drago has joined #openstack-ansible | 17:56 | |
| *** zerda2 has quit IRC | 17:59 | |
| openstackgerrit | Steve Lewis (stevelle) proposed openstack/openstack-ansible: Introduce a playbook for deploying Gnocchi https://review.openstack.org/324125 | 18:00 |
| *** KLevenstein has quit IRC | 18:01 | |
| *** KLevenstein has joined #openstack-ansible | 18:02 | |
| *** electrofelix has quit IRC | 18:02 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Set check_mode variable every time https://review.openstack.org/326664 | 18:05 |
| *** jorge_munoz has joined #openstack-ansible | 18:09 | |
| odyssey4me | automagically stevelle FYI https://review.openstack.org/326396 rebased | 18:12 |
| v1k0d3n | hey all. i have one stubborn container that refuses to build on an infrastructure host. has anyone seen this behavior? | 18:16 |
| *** tiagogomes has quit IRC | 18:16 | |
| v1k0d3n | it's only this one host that's receiving an error: SSH Error: data could not be sent to the remote host. | 18:17 |
| v1k0d3n | every other container on this box seems to be fine... | 18:17 |
| odyssey4me | v1k0d3n have you escalated the number of forks you're using? | 18:18 |
| odyssey4me | and have you verified that you're not out of disk space? | 18:19 |
| v1k0d3n | haven't played with that honestly. | 18:19 |
| v1k0d3n | think i found an issue... | 18:19 |
| *** weezS has joined #openstack-ansible | 18:19 | |
| v1k0d3n | even though i have an ip excluded...it seems as though this container has claimed the .1 of my net...(a router ip) | 18:19 |
| v1k0d3n | that's bad. | 18:19 |
| odyssey4me | oh really? I don't think I"ve seen that happen before. | 18:20 |
| odyssey4me | are your used_ip exclusions done right? | 18:20 |
| v1k0d3n | ah..hmm | 18:20 |
| v1k0d3n | might be the darn config. shoot. | 18:20 |
| v1k0d3n | took this deployment over and have corrected a lot. i guess i overlooked this one! | 18:21 |
| v1k0d3n | lol | 18:21 |
| *** admin0 has quit IRC | 18:21 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Remove Ironic container creation from AIO preparation https://review.openstack.org/326686 | 18:24 |
| v1k0d3n | so odyssey4me how can i "reassign" a container ip...is there an easy way to do this with the project? | 18:26 |
| v1k0d3n | i've excluded. i can destroy and rebuild the container too...worried it's going to use the same ip though. | 18:26 |
| *** ametts has quit IRC | 18:27 | |
| v1k0d3n | would that be automatically taken care of after a destroy and rebuild of the container, or would that hang around someplace? | 18:28 |
| odyssey4me | v1k0d3n if you destroy the container without removing it from the inventory or changing the inventory, it will always come back with the same config | 18:30 |
| odyssey4me | so destroy it, then modify the inventory (or remove it from the inventory), then rebuild it | 18:30 |
| odyssey4me | if it's something where you want the name to remain the same - then modify the inventory | 18:30 |
| v1k0d3n | ok, that's what i thought. is there any way to update the inventory for a single container/host? | 18:30 |
| odyssey4me | if you don't care about the name, you can remove it from the inventory | 18:31 |
| odyssey4me | the name is important if it's something that has registered as an agent for a service before | 18:31 |
| odyssey4me | if you want to remove it - use inventory-manage.py from the scripts/ directory | 18:31 |
| *** Ashana has quit IRC | 18:31 | |
| odyssey4me | if you want to edit it, I don't think the script can do that so you'll have to edit it by hand (after backing up of course) | 18:31 |
| odyssey4me | if you feel the urge to add an edit function into inventory-manage, please do :) | 18:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Switch Ironic role repo to use git.o.o https://review.openstack.org/326690 | 18:35 |
| *** admin0 has joined #openstack-ansible | 18:38 | |
| *** saneax_AFK is now known as saneax | 18:39 | |
| *** Zucan has quit IRC | 18:40 | |
| *** ametts has joined #openstack-ansible | 18:40 | |
| *** admin0 has quit IRC | 18:40 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Docs: Add developer guide for security role https://review.openstack.org/325499 | 18:45 |
| v1k0d3n | odyssey4me: yeah, this sounds like a good opportunity. at some point possibly...still trying to get my arms around my current workload. | 18:45 |
| *** weezS has quit IRC | 18:50 | |
| odyssey4me | ok, I'm out for the night - cheerio all! | 18:51 |
| *** admin0 has joined #openstack-ansible | 18:52 | |
| michaelgugino | cya | 18:52 |
| michaelgugino | If anyone that has knowledge of the tempest testing could take a look at https://review.openstack.org/#/c/324045/ I would appreciate it. | 18:56 |
| michaelgugino | Nova appears to install fine, and seems to be working locally, but the tempest functional test is complaining about some AZ not existing | 18:56 |
| *** mhayden has quit IRC | 19:00 | |
| *** mhayden has joined #openstack-ansible | 19:00 | |
| *** weezS has joined #openstack-ansible | 19:01 | |
| *** rahuls has joined #openstack-ansible | 19:04 | |
| *** galstrom is now known as galstrom_zzz | 19:07 | |
| *** galstrom_zzz is now known as galstrom | 19:08 | |
| *** evilt0ne has joined #openstack-ansible | 19:12 | |
| openstackgerrit | Mark Hooper proposed openstack/openstack-ansible-security: adding grub conf variable since RHEL 7 is now using /boot/grub2 instead of /boot/grub https://review.openstack.org/326705 | 19:14 |
| *** raddaoui has quit IRC | 19:17 | |
| openstackgerrit | Mark Hooper proposed openstack/openstack-ansible-security: forgot quotes, oops https://review.openstack.org/326709 | 19:21 |
| *** raddaoui has joined #openstack-ansible | 19:28 | |
| *** permalac has quit IRC | 19:36 | |
| *** asettle has quit IRC | 19:42 | |
| openstackgerrit | Mark Hooper proposed openstack/openstack-ansible-security: adding grub conf variable since RHEL 7 is now using /boot/grub2 instead of /boot/grub https://review.openstack.org/326705 | 19:44 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Fix grub configuration file path in RHEL/CentOS 7 https://review.openstack.org/326705 | 19:48 |
| mhayden | two new security role contributors in one week! | 19:49 |
| mhayden | automagically: time to celebrate! | 19:49 |
| automagically | I could use a beer | 19:49 |
| automagically | I’ll take that as an excuse | 19:49 |
| mhayden | woot | 19:51 |
| mhayden | automagically: mark is using the security role on RHEL 7 -- hence the patch i tossed into the security role for RHEL 7 support | 19:51 |
| mhayden | no way to gate that though, but the patch is small | 19:51 |
| automagically | Cool, I saw your update | 19:51 |
| *** admin0 has quit IRC | 19:59 | |
| *** asettle has joined #openstack-ansible | 19:59 | |
| *** galstrom is now known as galstrom_zzz | 20:03 | |
| *** admin0 has joined #openstack-ansible | 20:04 | |
| *** retreved has quit IRC | 20:05 | |
| *** galstrom_zzz is now known as galstrom | 20:32 | |
| *** BjoernT has quit IRC | 20:33 | |
| kboratynski | mhayden: I haven't seen STIGs for RHEL7. ;-/ | 20:33 |
| mhayden | kboratynski: me either :| | 20:33 |
| mhayden | then again, that opens the door to a lot more work :P | 20:34 |
| mhayden | so if they take their time a little more, i'm okay with that :P | 20:34 |
| kboratynski | mhayden: There are many test that might be useful, 'noexec' on /home, 'nodev' on /home, so on. ;-) | 20:35 |
| mhayden | kboratynski: if you have a moment to review some commits to the security repo, i'd be much obliged :) | 20:35 |
| mhayden | i'd like to get some solid logging so that audit mode could run fully without needing the fail module | 20:35 |
| kboratynski | mhayden: Me as a reviewer? :-) | 20:35 |
| mhayden | kboratynski: absolutely! why not? :) | 20:35 |
| kboratynski | Well, I'm ok with that! :-) | 20:35 |
| mhayden | anyone can provide input there | 20:36 |
| kboratynski | mhayden: I know 'reno' tool. Take it easy. ;-) I will look at your comments in 2 days time. | 20:36 |
| mhayden | but if you actually use the role, your input is super valuable | 20:36 |
| mhayden | haha, kboratynski -- that's cool, i know it's new to some folks | 20:36 |
| *** asettle has quit IRC | 20:36 | |
| kboratynski | mhayden: Yep. I use it. Not in OpenStack env, unfortunately, but yep. :-) | 20:37 |
| kboratynski | It confuses me, why there are no such security checks like provided above. Don't you think it is strange? | 20:39 |
| mhayden | how do you mean? | 20:40 |
| mhayden | automagically: was this change still good? https://review.openstack.org/289032 | 20:43 |
| kboratynski | mhayden: Sticky bit for /tmp. 'nodev', 'noexec' for /home as mountpoint options, ulimits, IDS and file integrity tool installed... and many others I haven't seen among STIGs. ;-( | 20:46 |
| mhayden | oh yeah, totally agree | 20:46 |
| mhayden | are you talking about host-based IDS? like OSSEC? | 20:46 |
| mhayden | i assume something more than file integrity | 20:46 |
| * mhayden hands stevelle a +2 for a big load of documentation :) | 20:48 | |
| stevelle | thank yas | 20:48 |
| kboratynski | host and network based IDS. | 20:48 |
| kboratynski | Depending on type of machine (host). | 20:48 |
| kboratynski | Something like Snort. | 20:48 |
| kboratynski | In example. | 20:48 |
| *** kylek3h has quit IRC | 20:54 | |
| *** asettle has joined #openstack-ansible | 20:59 | |
| *** saneax is now known as saneax_AFK | 20:59 | |
| *** Mudpuppy has quit IRC | 21:00 | |
| mhayden | that would be good | 21:02 |
| * mhayden is out for now | 21:02 | |
| mhayden | night folks | 21:02 |
| *** spotz is now known as spotz_zzz | 21:03 | |
| openstackgerrit | Merged openstack/openstack-ansible-security: Implemented: V-38524. https://review.openstack.org/324960 | 21:04 |
| *** goldenfri has joined #openstack-ansible | 21:09 | |
| *** psilvad has quit IRC | 21:09 | |
| *** thorst has quit IRC | 21:12 | |
| *** thorst has joined #openstack-ansible | 21:13 | |
| *** thorst has quit IRC | 21:17 | |
| *** pjm6 has quit IRC | 21:19 | |
| *** pjm6 has joined #openstack-ansible | 21:19 | |
| *** weezS has quit IRC | 21:27 | |
| *** thorst has joined #openstack-ansible | 21:38 | |
| *** admin0 has quit IRC | 21:41 | |
| *** thorst has quit IRC | 21:42 | |
| mrda | Morning all | 21:55 |
| *** cloudtrainme has joined #openstack-ansible | 21:56 | |
| *** messy has quit IRC | 21:57 | |
| *** spotz_zzz is now known as spotz | 21:58 | |
| *** KLevenstein has quit IRC | 21:58 | |
| openstackgerrit | Michael Davies proposed openstack/openstack-ansible: Add an easy way to run cmds in utility container https://review.openstack.org/326200 | 22:03 |
| *** asettle has quit IRC | 22:03 | |
| *** sdake_ has joined #openstack-ansible | 22:03 | |
| *** Nepoc has joined #openstack-ansible | 22:04 | |
| *** sdake has quit IRC | 22:06 | |
| lbragstad | question - is this evaluated as "and" or as an "or" in ansible? https://github.com/openstack/openstack-ansible-os_keystone/blob/master/meta/main.yml#L42-L44 | 22:06 |
| *** sdake has joined #openstack-ansible | 22:06 | |
| *** weshay has quit IRC | 22:08 | |
| openstackgerrit | Dolph Mathews proposed openstack/openstack-ansible-os_keystone: Always use memcached https://review.openstack.org/326748 | 22:08 |
| *** sdake_ has quit IRC | 22:10 | |
| openstackgerrit | Dolph Mathews proposed openstack/openstack-ansible-os_keystone: Always use memcached https://review.openstack.org/326748 | 22:11 |
| mattt | lbragstad: and | 22:24 |
| *** BjoernT has joined #openstack-ansible | 22:26 | |
| BjoernT | Hi folks, we just opened a critical issue https://bugs.launchpad.net/openstack-ansible/+bug/1590166. Galera instances recovery from SST is not possible due to crashing xtradb | 22:27 |
| openstack | Launchpad bug 1590166 in openstack-ansible "Rejoining/recovery of galera cluster fails" [Undecided,New] | 22:27 |
| *** fawadkhaliq has joined #openstack-ansible | 22:28 | |
| *** ametts has quit IRC | 22:30 | |
| *** weezS has joined #openstack-ansible | 22:39 | |
| *** cloudtrainme has quit IRC | 22:43 | |
| *** cloudtrainme has joined #openstack-ansible | 22:43 | |
| *** cloudtrainme has quit IRC | 22:43 | |
| *** sdake has quit IRC | 22:43 | |
| *** Drago has quit IRC | 22:45 | |
| *** pjm6_ has joined #openstack-ansible | 22:50 | |
| *** pjm6 has quit IRC | 22:52 | |
| *** pjm6 has joined #openstack-ansible | 22:57 | |
| openstackgerrit | Flávio Ramalho proposed openstack/openstack-ansible-os_sahara: Test sahara with tempest https://review.openstack.org/325673 | 22:58 |
| lbragstad | mattt ah - thanks! | 23:00 |
| lbragstad | cc dolphm ^ | 23:00 |
| *** pjm6_ has quit IRC | 23:00 | |
| mattt | lbragstad: np! | 23:02 |
| *** weezS has quit IRC | 23:02 | |
| *** psilvad has joined #openstack-ansible | 23:03 | |
| *** rahuls has quit IRC | 23:05 | |
| *** galstrom is now known as galstrom_zzz | 23:06 | |
| *** rahuls has joined #openstack-ansible | 23:09 | |
| *** sdake has joined #openstack-ansible | 23:12 | |
| *** markvoelker has quit IRC | 23:13 | |
| *** scarlisle has quit IRC | 23:13 | |
| *** sacharya_ has joined #openstack-ansible | 23:15 | |
| *** sacharya has quit IRC | 23:15 | |
| *** rahuls has quit IRC | 23:15 | |
| *** BjoernT has quit IRC | 23:16 | |
| *** spotz is now known as spotz_zzz | 23:17 | |
| *** saneax_AFK is now known as saneax | 23:22 | |
| *** sacharya_ has quit IRC | 23:25 | |
| *** rahuls has joined #openstack-ansible | 23:26 | |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-specs: Xen Virt Driver Support https://review.openstack.org/325490 | 23:26 |
| *** fawadkhaliq has quit IRC | 23:37 | |
| *** sdake has quit IRC | 23:46 | |
| *** fawadkhaliq has joined #openstack-ansible | 23:52 | |
| *** thorst has joined #openstack-ansible | 23:56 | |
| *** fawadkhaliq has quit IRC | 23:56 | |
| *** thorst has quit IRC | 23:56 | |
| *** thorst has joined #openstack-ansible | 23:57 | |
| *** saneax is now known as saneax_AFK | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!