| *** severion has joined #openstack-ansible | 00:01 | |
| *** v1k0d3n has quit IRC | 00:04 | |
| *** cmart has joined #openstack-ansible | 00:07 | |
| *** severion has quit IRC | 00:07 | |
| *** v1k0d3n has joined #openstack-ansible | 00:08 | |
| *** cmart has quit IRC | 00:09 | |
| *** cmart has joined #openstack-ansible | 00:09 | |
| *** dxiri has quit IRC | 00:12 | |
| *** smatzek has joined #openstack-ansible | 00:12 | |
| *** galstrom_zzz is now known as galstrom | 00:13 | |
| *** galstrom is now known as galstrom_zzz | 00:16 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_barbican: Update paste, policy and rootwrap configurations 2016-10-13 https://review.openstack.org/386276 | 00:18 |
|---|---|---|
| *** javeriak has joined #openstack-ansible | 00:19 | |
| *** javeriak has quit IRC | 00:20 | |
| *** cmart has quit IRC | 00:23 | |
| *** v1k0d3n has quit IRC | 00:28 | |
| *** v1k0d3n has joined #openstack-ansible | 00:28 | |
| *** javeriak has joined #openstack-ansible | 00:29 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Create minor playbook to sync TZ data https://review.openstack.org/386235 | 00:30 |
| *** smatzek has quit IRC | 00:32 | |
| *** smatzek has joined #openstack-ansible | 00:32 | |
| *** v1k0d3n has quit IRC | 00:33 | |
| *** xek_ has quit IRC | 00:35 | |
| *** xek_ has joined #openstack-ansible | 00:35 | |
| *** david-lyle_ has joined #openstack-ansible | 00:42 | |
| *** thorst has joined #openstack-ansible | 00:44 | |
| *** david-lyle has quit IRC | 00:45 | |
| *** thorst has quit IRC | 00:48 | |
| *** Mudpuppy_ has quit IRC | 00:48 | |
| *** Mudpuppy has joined #openstack-ansible | 00:48 | |
| *** evilt0ne has quit IRC | 00:52 | |
| *** Mudpuppy has quit IRC | 00:53 | |
| *** evilt0ne has joined #openstack-ansible | 00:54 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_barbican: Update paste, policy and rootwrap configurations 2016-10-13 https://review.openstack.org/386276 | 00:55 |
| openstackgerrit | zhongshengping proposed openstack/openstack-ansible-os_nova: Fix api_paste_config option https://review.openstack.org/386290 | 00:58 |
| *** vvro has quit IRC | 01:00 | |
| odyssey4me | FYI: http://lists.openstack.org/pipermail/openstack-dev/2016-October/105697.html | 01:06 |
| *** hughmFLEXin has joined #openstack-ansible | 01:09 | |
| *** shanec_ has joined #openstack-ansible | 01:12 | |
| *** dxiri has joined #openstack-ansible | 01:13 | |
| *** hughmFLEXin has quit IRC | 01:13 | |
| *** shanec__ has joined #openstack-ansible | 01:14 | |
| *** shanec_ has quit IRC | 01:16 | |
| *** dxiri has quit IRC | 01:17 | |
| *** gouthamr has quit IRC | 01:20 | |
| *** michaelgugino has quit IRC | 01:23 | |
| *** thorst has joined #openstack-ansible | 01:24 | |
| *** javeriak has quit IRC | 01:29 | |
| *** jamielennox is now known as jamielennox|away | 01:30 | |
| *** thorst has quit IRC | 01:33 | |
| *** jamielennox|away is now known as jamielennox | 01:37 | |
| *** retreved has quit IRC | 01:46 | |
| openstackgerrit | zhongshengping proposed openstack/openstack-ansible-os_heat: Deprecate auth_plugin option https://review.openstack.org/386301 | 01:46 |
| *** jcrst has quit IRC | 01:46 | |
| *** derekjhyang has joined #openstack-ansible | 01:50 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_barbican: Update paste, policy and rootwrap configurations 2016-10-13 https://review.openstack.org/386276 | 01:51 |
| *** sdake_ is now known as sdake | 02:01 | |
| *** dxiri has joined #openstack-ansible | 02:05 | |
| *** hughmFLEXin has joined #openstack-ansible | 02:05 | |
| *** rmelero has joined #openstack-ansible | 02:07 | |
| *** shanec__ has quit IRC | 02:11 | |
| *** galstrom_zzz is now known as galstrom | 02:15 | |
| *** adrian_otto has joined #openstack-ansible | 02:16 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_barbican: Update paste, policy and rootwrap configurations 2016-10-13 https://review.openstack.org/386276 | 02:21 |
| *** javeriak has joined #openstack-ansible | 02:21 | |
| *** adrian_otto has quit IRC | 02:40 | |
| *** rmelero has quit IRC | 02:45 | |
| *** adrian_otto has joined #openstack-ansible | 02:47 | |
| *** dxiri has quit IRC | 02:49 | |
| *** woodard has joined #openstack-ansible | 02:49 | |
| *** dxiri has joined #openstack-ansible | 02:49 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_barbican: Use barbican-manage https://review.openstack.org/386312 | 02:53 |
| *** adrian_otto has quit IRC | 02:57 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_barbican: Remove duplicate and unnecessary pip packages https://review.openstack.org/386318 | 03:20 |
| *** dxiri has quit IRC | 03:22 | |
| *** winggundamth has quit IRC | 03:27 | |
| *** dxiri has joined #openstack-ansible | 03:28 | |
| *** javeriak has quit IRC | 03:29 | |
| *** smatzek has quit IRC | 03:30 | |
| *** rstarmer has joined #openstack-ansible | 03:30 | |
| *** javeriak has joined #openstack-ansible | 03:32 | |
| *** hj-hpe has quit IRC | 03:36 | |
| *** omiday has quit IRC | 03:43 | |
| *** v1k0d3n has joined #openstack-ansible | 03:44 | |
| *** omiday has joined #openstack-ansible | 03:45 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts: Copy system localtime when creating container cache https://review.openstack.org/386187 | 03:54 |
| *** jcrst has joined #openstack-ansible | 03:56 | |
| *** rstarmer has quit IRC | 04:07 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift: Move flush handles to the end of the tasks https://review.openstack.org/386109 | 04:12 |
| *** hj-hpe has joined #openstack-ansible | 04:12 | |
| *** hj-hpe has quit IRC | 04:12 | |
| *** woodard has quit IRC | 04:15 | |
| *** david-lyle_ has quit IRC | 04:39 | |
| *** david-lyle has joined #openstack-ansible | 04:39 | |
| *** galstrom is now known as galstrom_zzz | 05:01 | |
| *** kamtamtun has quit IRC | 05:05 | |
| *** sdake has quit IRC | 05:23 | |
| dxiri | guys! quick question, how big should the container/mgmt network be? is a /24 enough? | 05:24 |
| dxiri | for a 7 or so nodes | 05:24 |
| *** Jack_Iv has joined #openstack-ansible | 05:28 | |
| *** v1k0d3n has quit IRC | 05:46 | |
| *** dxiri_ has joined #openstack-ansible | 06:05 | |
| *** dxiri has quit IRC | 06:08 | |
| *** dxiri_ has quit IRC | 06:09 | |
| *** Mudpuppy has joined #openstack-ansible | 06:21 | |
| *** pcaruana has joined #openstack-ansible | 06:23 | |
| *** Mudpuppy has quit IRC | 06:26 | |
| *** drifterza has joined #openstack-ansible | 06:42 | |
| drifterza | ehlo | 06:43 |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 06:51 |
| *** janki has joined #openstack-ansible | 06:56 | |
| *** asettle has joined #openstack-ansible | 07:01 | |
| *** asettle has quit IRC | 07:02 | |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 07:04 |
| *** admin0 has joined #openstack-ansible | 07:06 | |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 07:12 |
| *** rgogunskiy has joined #openstack-ansible | 07:14 | |
| *** Andrew_jedi has joined #openstack-ansible | 07:16 | |
| *** bsv has joined #openstack-ansible | 07:22 | |
| Mr_Smurf_ | Good morning | 07:23 |
| Mr_Smurf_ | I use liberty on ubuntu 14.04 deployed with openstack-ansible | 07:23 |
| Mr_Smurf_ | I get a strange error message in my cinder-api log and the error message "error unable to retrieve volume limit information" in horizon.. | 07:23 |
| Mr_Smurf_ | I have just switched from http to https for my public endpoints | 07:23 |
| Mr_Smurf_ | GET http://ssc-endpoint.hpc2n.umu.se:8776/v2/8f1fd12968a74fb59b2d266ce33ca89c/os-quota-sets/8f1fd12968a74fb59b2d266ce33ca89c?usage=False | 07:24 |
| Mr_Smurf_ | ERROR cinder.api.middleware.fault [req-386201b4-ded0-427b-ade6-02877a0a2f78 7f2dee0bc7b14e53a3e9f3415270887f 8f1fd12968a74fb59b2d266ce33ca89c - - -] Caught error: Could not determine a suitable URL for the plugin | 07:24 |
| Mr_Smurf_ | So I can understand the error message "INFO cinder.api.middleware.fault [req-386201b4-ded0-427b-ade6-02877a0a2f78 7f2dee0bc7b14e53a3e9f3415270887f 8f1fd12968a74fb59b2d266ce33ca89c - - -] http://ssc-endpoint.hpc2n.umu.se:8776/v2/8f1fd12968a74fb59b2d266ce33ca89c/os-quota-sets/8f1fd12968a74fb59b2d266ce33ca89c?usage=False returned with HTTP 500" | 07:24 |
| Mr_Smurf_ | since it request http insted of https | 07:24 |
| Mr_Smurf_ | But I cannot understand why it wants to use http since my endpoint list shows that the public endpoint is https://ssc-endpoint.hpc2n.umu.se and my internal endpint is http://172.16.2.1 (with some /v2 and /v1 stuff after) | 07:26 |
| *** javeriak has quit IRC | 07:34 | |
| Mr_Smurf_ | I only get the error when I look at network topology, I can create and delete volumes without any issues | 07:35 |
| *** karimb has joined #openstack-ansible | 07:38 | |
| evrardjp | good morning | 07:38 |
| *** allanice001 has joined #openstack-ansible | 07:39 | |
| *** gtrxcb has joined #openstack-ansible | 07:50 | |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 07:51 |
| admin0 | good morning all | 07:54 |
| bsv | admin0: o/ | 07:57 |
| admin0 | ello | 07:58 |
| admin0 | \o | 07:58 |
| drifterza | o/ | 07:58 |
| evrardjp | good morning all | 08:04 |
| allanice001 | morning all | 08:04 |
| *** gtrxcb has quit IRC | 08:10 | |
| admin0 | instead of doing ssh <controller> lxc-ls -f .. and then ssh to the ip , or ssh to the ssh controller ; lxc-attach -n <name> ; could there be a better way to login directly to the containers ? | 08:11 |
| allanice001 | admin0: ansible should have the inventory, and ssh ability, so tie into the existing inventory | 08:13 |
| *** haad1 has joined #openstack-ansible | 08:14 | |
| evrardjp | allanice001: if you want to have a friendly dump of the inventory | 08:14 |
| evrardjp | cd /opt/openstack-ansible | 08:14 |
| evrardjp | scripts/inventory-manage -l | 08:15 |
| evrardjp | .py* | 08:15 |
| evrardjp | it will list the generated containers, their group and their ips | 08:15 |
| evrardjp | so you can ssh manually into them | 08:15 |
| *** Jeffrey4l_ has quit IRC | 08:16 | |
| evrardjp | else you can use ansible | 08:16 |
| bsv | admin0: i did a script for that, using a controlmaster and a simple python script. | 08:16 |
| evrardjp | just go into your playbooks folder and it's gonna be simple | 08:16 |
| evrardjp | admin0: all of my text was in fact for you :p | 08:16 |
| allanice001 | :P | 08:17 |
| admin0 | yesssir :) | 08:17 |
| admin0 | aye aye cptain | 08:17 |
| admin0 | got it :) | 08:17 |
| evrardjp | allanice001: sorry for the spam :p | 08:17 |
| allanice001 | np evrardjp | 08:17 |
| admin0 | \o/ | 08:18 |
| admin0 | thanks evrardjp | 08:18 |
| admin0 | exactly what i was looking for | 08:18 |
| evrardjp | yw | 08:18 |
| allanice001 | admin0: you could seperate investigate teleport nodes as port of your deployment - http://gravitational.com/teleport | 08:18 |
| bsv | evrardjp: i loaded that info into a list in python and used it for the auto-complete feature of cmd-module. That way it was very fast to move around. | 08:18 |
| allanice001 | but i beleive the ansible method would be the fastes, as the inventory will always be up to date | 08:20 |
| evrardjp | bsv hehe, did you share it somewhere? | 08:20 |
| evrardjp | you should probably | 08:20 |
| drifterza | haad1, you around dude ? | 08:20 |
| admin0 | i forgot .. how do I source the existing ansible :D | 08:21 |
| admin0 | i meant the openstack ansible | 08:21 |
| admin0 | nvm.. will check this chat log | 08:21 |
| drifterza | . /usr/local/bin/openstack-ansible.rc | 08:21 |
| bsv | evrardjp: i can check later and put it on github. | 08:22 |
| admin0 | thanks .. the path was what i was missing :) | 08:22 |
| drifterza | np | 08:22 |
| evrardjp | bsv it's not really for me, so don't take too much effort into it | 08:22 |
| evrardjp | but I guess some ppl could love it | 08:22 |
| *** asettle has joined #openstack-ansible | 08:23 | |
| bsv | kk | 08:23 |
| *** fops has quit IRC | 08:25 | |
| *** fops has joined #openstack-ansible | 08:26 | |
| *** Jeffrey4l_ has joined #openstack-ansible | 08:43 | |
| *** Andrew_jedi has quit IRC | 08:48 | |
| *** jcrst has quit IRC | 08:58 | |
| *** jcrst has joined #openstack-ansible | 09:00 | |
| *** Andrew_jedi has joined #openstack-ansible | 09:04 | |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 09:09 |
| *** berendt has joined #openstack-ansible | 09:18 | |
| *** haad1 has quit IRC | 09:24 | |
| *** haad1 has joined #openstack-ansible | 09:33 | |
| *** evilt0ne has quit IRC | 09:36 | |
| *** evilt0ne has joined #openstack-ansible | 09:37 | |
| *** evilt0ne has quit IRC | 09:37 | |
| *** evilt0ne has joined #openstack-ansible | 09:37 | |
| *** openstackstatus has quit IRC | 09:44 | |
| *** openstack has joined #openstack-ansible | 09:48 | |
| *** jcrst has quit IRC | 09:48 | |
| *** v1k0d3n has joined #openstack-ansible | 09:48 | |
| *** v1k0d3n has quit IRC | 09:53 | |
| odyssey4me | bsv yeah, that'd be a really nifty thing to have :) | 09:55 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts: Copy system localtime when creating container cache https://review.openstack.org/386517 | 10:06 |
| *** haad1 has quit IRC | 10:10 | |
| *** openstackstatus has quit IRC | 10:13 | |
| *** openstack has joined #openstack-ansible | 10:16 | |
| odyssey4me | for anyone who has the inclination to get this done before Tuesday, I think this should be included in the Newton release: https://bugs.launchpad.net/openstack-ansible/+bug/1633411 | 10:22 |
| openstack | Launchpad bug 1633411 in openstack-ansible "Automate the sourcing of openstack-ansible.rc" [Wishlist,New] | 10:22 |
| *** Mudpuppy has joined #openstack-ansible | 10:22 | |
| *** smatzek has joined #openstack-ansible | 10:25 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-tests: Fix neutron_agent IP to be neutron_server IP https://review.openstack.org/386530 | 10:26 |
| *** Mudpuppy has quit IRC | 10:27 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_neutron: [WIP] Test Calico testing https://review.openstack.org/386526 | 10:40 |
| *** fxpester has joined #openstack-ansible | 10:43 | |
| *** pester has quit IRC | 10:45 | |
| *** haad1 has joined #openstack-ansible | 10:51 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: Add full path to inventory https://review.openstack.org/386542 | 11:06 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: Add full path to inventory https://review.openstack.org/386542 | 11:07 |
| evrardjp | odyssey4me: ^ if you want this to behave the same way on upgrade, we can simply remove the condition | 11:09 |
| andymccr | sweet calico testing passing: https://review.openstack.org/#/c/386526/ | 11:09 |
| evrardjp | oh and it's not tested | 11:09 |
| evrardjp | cool andymccr | 11:09 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_neutron: Fix Calico testing https://review.openstack.org/386526 | 11:10 |
| odyssey4me | evrardjp andymccr please help with tagging any bugs with the tag 'newton-rc-potential' to make it easier to find them | 11:10 |
| odyssey4me | any bugs that are important for the release that is | 11:11 |
| evrardjp | ok will check everything new | 11:11 |
| evrardjp | I tag what I think should be evaluated for newton but not yet triaged, so there could be garbage in it | 11:14 |
| evrardjp | we'll evaluate and remove if necessary | 11:14 |
| odyssey4me | agreed | 11:16 |
| odyssey4me | unfortunately our final tag deadline for an RC is Tuesday, so we'll need to triage and fix before COB Tue | 11:17 |
| evrardjp | ok | 11:17 |
| evrardjp | odyssey4me: when was this done? https://bugs.launchpad.net/openstack-ansible/+bug/1632831 | 11:17 |
| openstack | Launchpad bug 1632831 in openstack-ansible "The rabbitmq deterministic sort is now pointless" [Undecided,New] | 11:17 |
| odyssey4me | I'm assigning importance based on what I see in the bug breaks, but I'm not confirming just yet. | 11:17 |
| evrardjp | I mean the oslo part | 11:17 |
| evrardjp | is that recent? | 11:17 |
| evrardjp | fine for me | 11:18 |
| odyssey4me | evrardjp Nope - that change was made back in Juno/Kilo. | 11:18 |
| evrardjp | ok so we should carry this but low importance | 11:18 |
| odyssey4me | yep | 11:18 |
| evrardjp | it's a change of feature 'though, so it's not really something we should carry in newton | 11:18 |
| evrardjp | so except if you really want it for newton, I suggest postpone | 11:19 |
| odyssey4me | nah, it's not urgent | 11:19 |
| odyssey4me | I just wanted to note it as a bug so that we do something about it at some point | 11:19 |
| odyssey4me | it's an easy thing to fix and even backport | 11:19 |
| odyssey4me | what we care about now are more critical things like usability, incorrect docs or bugs that break things | 11:20 |
| evrardjp | yes that was my understanding | 11:20 |
| evrardjp | I'm now checking current confirmed/in progress/triaged for N | 11:21 |
| *** rgogunskiy has quit IRC | 11:21 | |
| *** ChanServ changes topic to "Bug smash list for Newton release: https://goo.gl/FOwKxC || Launchpad: https://launchpad.net/openstack-ansible || Weekly Meetings: https://wiki.openstack.org/wiki/Meetings/openstack-ansible || Review Dashboard: https://goo.gl/tTmdgs" | 11:23 | |
| odyssey4me | ok, I've reversed the sort order to make those marked critical/high at the top of the list | 11:24 |
| odyssey4me | note that the ones that are 'incomplete' are the ones which have a response | 11:25 |
| odyssey4me | it looks like some of them can be closed as invalid because there are reports of not being able to replicate the issue | 11:25 |
| odyssey4me | I'll leave that to you evrardjp if that's ok? | 11:25 |
| evrardjp | yup fine for me | 11:26 |
| *** batkins61 has joined #openstack-ansible | 11:26 | |
| evrardjp | I finish the first round of tagging | 11:26 |
| *** dgonzalez has quit IRC | 11:26 | |
| evrardjp | this way I'm not lost with my open tabs | 11:26 |
| logan- | Thanks andymccr! | 11:27 |
| andymccr | logan-: np! just need to merge the depends on PR for the test repo, then that one should pass again | 11:27 |
| odyssey4me | andymccr we should probably setup a blueprint for M->N upgrades and associate all the bugs related ot that topic to the blueprint so that they're easier to find | 11:29 |
| odyssey4me | either that or tag them | 11:29 |
| odyssey4me | not sure how you want to go with that | 11:29 |
| *** thorst_ has joined #openstack-ansible | 11:32 | |
| evrardjp | fair point | 11:35 |
| *** retreved has joined #openstack-ansible | 11:38 | |
| *** batkins61_ has joined #openstack-ansible | 11:39 | |
| evrardjp | odyssey4me: for https://review.openstack.org/#/c/386542 what's the missing part? | 11:39 |
| evrardjp | you don't plan to have a wrapper around all the ansible commands, do you? | 11:40 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: Add full path to inventory https://review.openstack.org/386542 | 11:41 |
| *** batkins61 has quit IRC | 11:41 | |
| *** batkins61_ is now known as batkins61 | 11:42 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: Add full path to inventory https://review.openstack.org/386542 | 11:42 |
| odyssey4me | evrardjp we already kinda do | 11:42 |
| evrardjp | for playbooks | 11:42 |
| evrardjp | and you want to extend for ad-hoc commands that's what you mean in the bug? | 11:43 |
| evrardjp | Not sure to understand :p | 11:43 |
| odyssey4me | evrardjp see https://github.com/openstack/openstack-ansible/blob/master/scripts/bootstrap-ansible.sh#L109-L112 | 11:43 |
| *** fops_ has joined #openstack-ansible | 11:43 | |
| *** dgonzalez has joined #openstack-ansible | 11:43 | |
| *** karimb has quit IRC | 11:44 | |
| evrardjp | so I understood you ... you want a wrapper for everything... well I'll remove my assignment from this | 11:44 |
| odyssey4me | lol, I'm open to suggestion - the problem is real - the question is how we deal with it | 11:45 |
| *** fops has quit IRC | 11:45 | |
| odyssey4me | I suspect in an upgrade situation we may end up with conflicting Ansible version too, which isn't good | 11:46 |
| evrardjp | IMO, creating a wrapper is fine, but tricking users with a different behavior for the same command by default is not | 11:46 |
| odyssey4me | lemme register a bug for that | 11:46 |
| evrardjp | I'll handle the bug list now to mark bugs as invalid/incomplete | 11:47 |
| evrardjp | admin0: https://bugs.launchpad.net/openstack-ansible/+bug/1630948 | 11:52 |
| openstack | Launchpad bug 1630948 in openstack-ansible "keepalive does not work from the start" [High,New] | 11:52 |
| admin0 | well, what i have found is that in multi-node installs, even though the installation is done, the IP does not come up | 11:53 |
| admin0 | and i had to login and manually start keepalive in random (hoping) one of the 3 will pickup | 11:53 |
| admin0 | which it does | 11:53 |
| admin0 | “The VIP could take a few moments to be spawned up with the election” — how much is the few momments here ? | 11:54 |
| *** rgogunskiy has joined #openstack-ansible | 11:54 | |
| odyssey4me | admin0 and if you use an IP instead of a domain name for the internal address? | 11:55 |
| evrardjp | admin0: it depends | 11:55 |
| *** karimb has joined #openstack-ansible | 11:55 | |
| evrardjp | admin0: of your checks/processes/network | 11:55 |
| admin0 | the domains are resolveable .. they all point to address that can be reached and resolved | 11:56 |
| admin0 | odyssey4me: in production i would use the domain name( instead of IP) so i only checked via domain name | 11:56 |
| bsv | always domains | 11:57 |
| odyssey4me | admin0 yes, but I'm asking you to try it as a way of determining the source of the problem - it's a triage step | 11:57 |
| evrardjp | odyssey4me: hostname is not relevant for keepalived | 11:57 |
| evrardjp | pingability of the default address could be | 11:57 |
| odyssey4me | and FYI, it's porbably better to use IP's for the internal endpoints and a domain name for the external endpoint | 11:57 |
| odyssey4me | the internal endpoints are used by systems - not people | 11:58 |
| evrardjp | can't find keepalived documentation in new guide :/ | 11:58 |
| odyssey4me | evrardjp http://docs.openstack.org/developer/openstack-ansible-haproxy_server/ | 11:58 |
| *** johnmilton has joined #openstack-ansible | 11:58 | |
| admin0 | odyssey4me, we have customers whom we have given internal endpoints to use for some speciifc use cases | 11:58 |
| *** johnmilton has quit IRC | 11:58 | |
| evrardjp | admin0: http://docs.openstack.org/developer/openstack-ansible-haproxy_server/configure-haproxy.html#configuring-keepalived-ping-checks | 11:59 |
| evrardjp | thanks odyssey4me | 11:59 |
| admin0 | but again, true that is bad .. | 11:59 |
| odyssey4me | admin0 that is somewhat frightening, but hey - your business not mine :p | 11:59 |
| evrardjp | admin0: could your keepalived hosts ping 193.0.14.129 | 12:00 |
| *** johnmilton has joined #openstack-ansible | 12:00 | |
| evrardjp | are the nics fine | 12:00 |
| evrardjp | these are my questions | 12:00 |
| admin0 | they can | 12:00 |
| bsv | odyssey4me: when you move to ipv6, not using dns starts to hurt. | 12:00 |
| evrardjp | yup that's what I didn't go that far by discarding internal names for lb :p | 12:01 |
| evrardjp | but anyway that's another topic | 12:01 |
| evrardjp | let's focus on keepalived | 12:01 |
| *** rgogunskiy has quit IRC | 12:01 | |
| admin0 | i can break down the current env and retry | 12:01 |
| admin0 | or in an hour, i will have a new 4 node env to play with .. can try there as well | 12:02 |
| admin0 | regarding the https://bugs.launchpad.net/openstack-ansible/+bug/1630948 and the Ips i use, cloud101.stack31.com does map to 10.11.12.3 (internally) because there is a different dns system for internal that gives it an internal ip | 12:04 |
| openstack | Launchpad bug 1630948 in openstack-ansible "keepalive does not work from the start" [High,New] | 12:04 |
| admin0 | so dns is not an issue | 12:04 |
| evrardjp | it never was | 12:04 |
| odyssey4me | bsv point taken :) | 12:04 |
| evrardjp | keepalived process dying IS a problem | 12:04 |
| odyssey4me | now if only rabbitmq would support IPv6 | 12:05 |
| evrardjp | not to be taken lightly | 12:05 |
| admin0 | onlything in the log = 2016-10-05 21:59:40 Unknown state | 12:05 |
| admin0 | not dying i guess .. not working | 12:05 |
| evrardjp | odyssey4me: I think it can | 12:05 |
| admin0 | for which i had to restart | 12:05 |
| admin0 | all 3 keepalived in logs were unknown state | 12:05 |
| evrardjp | oh | 12:05 |
| evrardjp | that's different | 12:05 |
| admin0 | and a quick reboot of either one ( or all) fixed it | 12:05 |
| evrardjp | process are up | 12:05 |
| odyssey4me | orly? last I checked it couldn't - and that's been a blocker for us switching the control plane to IPv6 | 12:05 |
| evrardjp | I think it's quite recent | 12:05 |
| evrardjp | maybe not stable yet | 12:06 |
| admin0 | sorry i should have mentioned the bug clearly .. i do not think i found the process dead | 12:06 |
| evrardjp | I have to check | 12:06 |
| admin0 | just not working | 12:06 |
| odyssey4me | also, IIRC logan- or someone pushed up patches to allow haproxy to run in containers - we should probably switch to that as a default :) | 12:06 |
| admin0 | and even after 5 mins of waiting, no VIP address working and logs only show 1 line ; unknown state | 12:06 |
| evrardjp | admin0: well the IP is not gonna be up until someone is ready to be master | 12:06 |
| evrardjp | for that you have to check what's wrong in your env | 12:06 |
| evrardjp | ping issues | 12:06 |
| bsv | odyssey4me: [ | 12:06 |
| bsv | {rabbit, [ | 12:06 |
| bsv | {tcp_listeners, [{"127.0.0.1", 5672}, | 12:06 |
| bsv | {"::1", 5672}]} | 12:06 |
| bsv | ]} | 12:06 |
| evrardjp | multicast issues | 12:06 |
| bsv | ].:) | 12:06 |
| odyssey4me | heh, according to http://www.rabbitmq.com/networking.html ipv6 is now a thing | 12:06 |
| admin0 | evrardjp: no ping issues | 12:07 |
| bsv | yep | 12:07 |
| admin0 | networking is sane | 12:07 |
| admin0 | a restart of one or more automatically promotes a master | 12:07 |
| evrardjp | well that's normal | 12:07 |
| admin0 | yes | 12:07 |
| admin0 | that is true as well | 12:07 |
| evrardjp | I mean | 12:07 |
| odyssey4me | according to https://linuxcontainers.org/lxc/manpages/man5/lxc.container.conf.5.html lxc containers can too | 12:08 |
| admin0 | but the thing is, this is done via the setup-infra .. and later int he process it fails because the repo proxies are not found .. | 12:08 |
| odyssey4me | heh, now all we need is someone to take on the work :) | 12:08 |
| admin0 | and to rerun, you have to delete the 00apt-proxy | 12:08 |
| evrardjp | if you have on 3 islands parts of an archipel they talk to each other everything is fine | 12:08 |
| evrardjp | then one decides to brexit | 12:08 |
| admin0 | and rerun again .. which does not restart haproxy | 12:08 |
| admin0 | :D | 12:08 |
| odyssey4me | admin0 if you use the latest newton tag, you won't have to delete the apt-proxy config | 12:08 |
| evrardjp | well when you brexit you're alone on your island and it works fine if you don't check others | 12:09 |
| evrardjp | but that doesn't mean it's fine | 12:09 |
| admin0 | that is what they said just go to get the votes :D | 12:09 |
| evrardjp | so | 12:09 |
| evrardjp | one question | 12:09 |
| evrardjp | are your haproxy processes up | 12:09 |
| admin0 | right now everything is UP | 12:10 |
| evrardjp | kill a node | 12:10 |
| admin0 | i need to break that env, or retry it in a new env | 12:10 |
| evrardjp | kill a haproxy for example | 12:10 |
| evrardjp | or cut the nic | 12:10 |
| admin0 | can you PMme our ssh key, and you can do what you want :D | 12:10 |
| admin0 | your* | 12:10 |
| evrardjp | I don't think there is a bug | 12:10 |
| evrardjp | :p | 12:11 |
| evrardjp | maybe a race condition | 12:11 |
| *** karimb has quit IRC | 12:11 | |
| admin0 | this step stops a smooth setup-infra everytime | 12:11 |
| evrardjp | or if there is a bug, it's with something else than keepalived | 12:11 |
| evrardjp | admin0: ok let's make this simple | 12:11 |
| odyssey4me | andymccr jmccrory chris_hultin|AWA I was thinking about the magnum role, and the long test time. Perhaps a way to solve it is to ditch the LXC container deployment from the test entirely and just do the install on the host only? | 12:11 |
| *** Andrew_jedi has quit IRC | 12:12 | |
| evrardjp | could you reproduce it 100% of the time? | 12:12 |
| odyssey4me | It'll cut some time from the deployment. | 12:12 |
| evrardjp | if yes, before fixing anything, could you drop my key there? | 12:12 |
| admin0 | evrardjp: | 12:12 |
| admin0 | let me destroy this env | 12:12 |
| admin0 | and then do it again | 12:12 |
| admin0 | in that way, i can reproduce it | 12:12 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_ironic: [WIP] Enable virtualBMC testing https://review.openstack.org/381049 | 12:13 |
| admin0 | drifterza was logged in there as well when we built it up .. so this was when decided to file for the bug report because it was encountered more than 3-4 times in a row | 12:13 |
| admin0 | actually, my new infra will be up in approx 30 mins . 4 node system | 12:15 |
| admin0 | will try it there | 12:15 |
| mgariepy | good morning everyone | 12:16 |
| admin0 | drifterza: you around ? | 12:16 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-haproxy_server: Fix markup error https://review.openstack.org/386562 | 12:17 |
| *** kylek3h has joined #openstack-ansible | 12:25 | |
| *** kamtamtun has quit IRC | 12:25 | |
| *** gouthamr has joined #openstack-ansible | 12:26 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: [WIP] Attempt to fix swift_pypy memory issues https://review.openstack.org/386566 | 12:27 |
| *** bsv has quit IRC | 12:28 | |
| *** johnmilton has quit IRC | 12:33 | |
| *** johnmilton has joined #openstack-ansible | 12:34 | |
| admin0 | evrardjp: i will run this again on 2 different new greenfield setups and update again | 12:34 |
| *** johnmilton has quit IRC | 12:34 | |
| ioni | does anybody know the freenode channel to talk abot gnocchi problems? | 12:34 |
| ioni | is #openstack-telemetry? | 12:35 |
| *** johnmilton has joined #openstack-ansible | 12:35 | |
| odyssey4me | ioni probably | 12:35 |
| *** karimb has joined #openstack-ansible | 12:35 | |
| *** johnmilton has quit IRC | 12:35 | |
| *** johnmilton has joined #openstack-ansible | 12:35 | |
| evrardjp | admin0: cool | 12:36 |
| *** openstackstatus has quit IRC | 12:43 | |
| *** openstack has joined #openstack-ansible | 12:46 | |
| *** cristicalin has joined #openstack-ansible | 12:48 | |
| *** fguillot has joined #openstack-ansible | 12:50 | |
| admin0 | TASK [os_keystone : Ensure newest key is used for credential in Keystone] ****** failing there ( this is to check if i can drop a mitaka database and run the os-keystone playbook agian to check if it upgrades | 12:50 |
| admin0 | trying to install it as region2 on an existing keystone db | 12:55 |
| *** Andrew_jedi has joined #openstack-ansible | 12:58 | |
| admin0 | new bug title: openstack-ansible does not install as region2 on an existing keystone db :D | 12:59 |
| odyssey4me | admin0 no, auth_token is in keystone.conf if you configure it - it was deprecated in Mitaka and is a horrible security hole | 12:59 |
| admin0 | need some help/pointers | 12:59 |
| odyssey4me | admin0 we had a discussion last night about doing that - see http://eavesdrop.openstack.org/irclogs/%23openstack-ansible/%23openstack-ansible.2016-10-13.log.html#t2016-10-13T22:15:46 | 12:59 |
| odyssey4me | there's a bit of chat before and after that, noting a sample config and some caveats | 13:00 |
| logan- | admin0: you probably will want to get rid of identity_hosts from your inventory if you're interfacing with a separate keystone | 13:00 |
| admin0 | no no .. if we do that, the legacy keystone will always remain | 13:00 |
| admin0 | the thing i am trying is this | 13:01 |
| admin0 | 1. setup openstack newton, but as region: region2 .. so i check endpoints and all are registered as region2 | 13:01 |
| logan- | ah | 13:01 |
| logan- | service_region: region2 is all it should take to do that | 13:01 |
| admin0 | then i drop the keystone db , and import a mitaka keystone db from region 1 .. so hope is.. os-keystone will check its existing db, upgade it from mitaka to newton and then install the endpoints ( but as region2 ) .. since i am using different usernames for service like keystone2, newton2 etc .. it should just work :D .. and then i can just point to this new endpoitns from existing region1 and retire the keystone there .. and have | 13:02 |
| admin0 | this shiny new ansible managed keystone | 13:02 |
| admin0 | thats the plan | 13:02 |
| *** sdake_ has joined #openstack-ansible | 13:03 | |
| admin0 | first roadblock is TASK [os_keystone : Ensure newest key is used for credential in Keystone — stuck there :) | 13:03 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: [TESTING] Checking CI jobs https://review.openstack.org/386582 | 13:04 |
| admin0 | so need help from someone who can get this installed in the existing kesytone db .. .. so that ultimately, the old cluster/region can retire | 13:05 |
| admin0 | using multilple keystone, or keystone auth to region1 keystone = keeping legacy | 13:05 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for 13.3.6 https://review.openstack.org/386584 | 13:05 |
| openstackgerrit | Frédéric Guillot proposed openstack/openstack-ansible-os_almanach: [WIP] Make project compatible with CI infra https://review.openstack.org/386233 | 13:05 |
| *** jamesdenton has joined #openstack-ansible | 13:06 | |
| admin0 | if this works we mortals can easily deply a new region ( where the new region always has the latest keystone)/db, and then easy to retire old regions | 13:06 |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 13:07 |
| admin0 | odyssey4me: who is going to look into this ? how can i help ? | 13:07 |
| admin0 | i mean the outcome of the talks int he chat from yesterday | 13:07 |
| odyssey4me | admin0 to get that right you need to prep the old DB to ensure that it has the credentials for the new environment, then you need to ensure that the user_secrets for the new environment match the creds needed to access the services and authenticate | 13:08 |
| odyssey4me | admin0 the point of the discussion is that all the tools are there - only you have to figure out how to use them for you use case | 13:08 |
| odyssey4me | no-one is going to do your work for you | 13:08 |
| admin0 | just need pointers :D | 13:08 |
| odyssey4me | admin0 you have many, many pointers - you've been pointed at the same things over and over again | 13:09 |
| odyssey4me | we can point, but you have to try and figure it out | 13:09 |
| * bsv points towards east... | 13:09 | |
| admin0 | and still if i fail after multiple retries, then i need some help :) | 13:09 |
| odyssey4me | yes, then do some digging, ask the questions and take the advice you're given | 13:09 |
| *** cristicalin has quit IRC | 13:10 | |
| admin0 | odyssey4me: “ prep the old DB to ensure that it has the credentials for the new environment” — stuck there :(( | 13:10 |
| odyssey4me | this is an open source project - it's all open for you to read - the existing contributors and users can help and guide you a bit, but you also have to take on the task of following the logic and figuring out where your use case is failing | 13:11 |
| *** jheroux has joined #openstack-ansible | 13:11 | |
| odyssey4me | admin0 if you want to use different usernames, passwords, etc for the database then you need to add them before you move it | 13:11 |
| odyssey4me | also, you need to ensure that the service user can even access the database | 13:11 |
| odyssey4me | either that or the user_secrets values for all the services must match those of the old environment | 13:11 |
| odyssey4me | it's fairly logical if you think about it | 13:12 |
| odyssey4me | how do the services authenticate or access the DB if they don'thave the rights to do so in an existing DB | 13:12 |
| *** cathrichardson has joined #openstack-ansible | 13:13 | |
| *** cathrichardson has quit IRC | 13:13 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for 12.2.6 https://review.openstack.org/386591 | 13:13 |
| *** v1k0d3n has joined #openstack-ansible | 13:14 | |
| *** cathrichardson has joined #openstack-ansible | 13:14 | |
| admin0 | odyssey4me i setup a greenfield neuton .. then just drop keystone db and restore from mitaka .. so in this case, all access passwords and usernames are the same .. what i am trying to get is have the keystone playbook upgrade this db and add the new endpoints as region2 .. because there is no region2 entry, it should just add right ? .. the services are still in newton for this region ( and noother database is touched ) .. | 13:14 |
| admin0 | region1 still maintains its own db .. just that the keystone API needs to be pointed to this new env | 13:14 |
| admin0 | so i am a bit lost about the user/pas you mentioned | 13:15 |
| openstackgerrit | Frédéric Guillot proposed openstack/openstack-ansible-os_almanach: [WIP] Make project compatible with CI infra https://review.openstack.org/386233 | 13:15 |
| admin0 | keystone can still access the db without change ( beause just the db content changed for keystone, not the mysql) | 13:15 |
| *** jperry has joined #openstack-ansible | 13:15 | |
| odyssey4me | admin0 ok, how about you first do it without changing the region? validate that everything works - especially the creds | 13:16 |
| logan- | the keystone playbook doesn't add endpoints for other services. | 13:16 |
| logan- | nova playbook adds the nova endpoints etc. | 13:16 |
| odyssey4me | and yes, logan- is right - each service does its own registration | 13:16 |
| odyssey4me | also, you can simply register any endpoints you want manually | 13:16 |
| odyssey4me | openstack endpoint add ... | 13:17 |
| odyssey4me | or probably create :p | 13:17 |
| admin0 | that is correct .. but the user case here is to retain the old keystone data and have the playbooks upgrade/do it to be future safe | 13:17 |
| admin0 | i will give it a few more tries | 13:17 |
| odyssey4me | admin0 another issue with your approach - you're replacing the keystone DB - which means that all the service setups for all the other services will be broken as all the uuid's will have changed | 13:18 |
| openstackgerrit | Kyle L. Henderson proposed openstack/openstack-ansible-galera_server: Move systemd handler above restart handler https://review.openstack.org/386604 | 13:18 |
| odyssey4me | it would be better if you put the DB in before doing the initial build of the other services | 13:18 |
| admin0 | if i rerun the respective playbooks aagain, won’t those be recreated ? | 13:18 |
| odyssey4me | ie complete setup-hosts and setup-infrastructure, then drop in the keystone DB, then run setup-openstack | 13:18 |
| odyssey4me | admin0 no - once the endpoints are created we specifically don't touch them | 13:19 |
| odyssey4me | this is to prevent unwanted changes later if, say, you edited them yourself | 13:19 |
| admin0 | ok .. btu in this case. since there are no region2 entries there anyway, its going to create it | 13:19 |
| admin0 | “ie complete setup-hosts and setup-infrastructure, then drop in the keystone DB, then run setup-openstack” — i will try this method | 13:20 |
| odyssey4me | kylek3h nice catch that | 13:20 |
| odyssey4me | we should actually audit all the handlers to make sure they're in the right order | 13:20 |
| kylek3h | odyssey4me: yeah, i wondered if there are more instances. | 13:20 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for 12.2.6 https://review.openstack.org/386591 | 13:23 |
| odyssey4me | andymccr evrardjp what do we think about backporting the dict-based service implementations to newton? | 13:24 |
| evrardjp | not sure what you mean | 13:25 |
| openstackgerrit | Frédéric Guillot proposed openstack/openstack-ansible-os_almanach: [WIP] Make project compatible with CI infra https://review.openstack.org/386233 | 13:25 |
| evrardjp | like the neutron way and the thing we discussed with groups? | 13:26 |
| admin0 | evrardjp: i am redoing the 3 api nodes to do the multi-region stuff like odyssey4me suggested .. will update how keepalive works this time | 13:27 |
| evrardjp | probably a community discussion? | 13:27 |
| evrardjp | admin0: ok | 13:28 |
| evrardjp | you install keepalived through the haproxy playbook right? | 13:28 |
| admin0 | yes | 13:28 |
| evrardjp | with the proper user variables like explained in the bug? | 13:28 |
| evrardjp | ok | 13:28 |
| evrardjp | cleanup your conf.d before maybe | 13:28 |
| evrardjp | haproxy confd | 13:28 |
| admin0 | the 3 controllers are having a total rebuilt | 13:29 |
| admin0 | none of the old stuff wil be there | 13:29 |
| *** jroll is now known as jrollinhatin | 13:29 | |
| admin0 | the deploy will also be clear of all old variables and inventory | 13:29 |
| evrardjp | ok | 13:29 |
| evrardjp | clean then | 13:29 |
| evrardjp | perfect | 13:29 |
| drifterza | odyssey4me +2 on dict-based backport | 13:30 |
| admin0 | so once the system is up, i will do setup-hosts followed by haproxy-setup | 13:30 |
| admin0 | before running setup-infra | 13:30 |
| evrardjp | drifterza: I'd like to make sure I'm talking about the same thing | 13:30 |
| cloudnull | mornings | 13:31 |
| evrardjp | morning cloudnull | 13:31 |
| drifterza | evrardjp I don't follow | 13:31 |
| admin0 | odyssey4me: existing db_version is 96 .. should be good ? | 13:32 |
| odyssey4me | admin0 no idea - you'll find out soon enough | 13:32 |
| admin0 | :D | 13:32 |
| admin0 | yep | 13:32 |
| evrardjp | odyssey4me: are we talking about some things like this? https://review.openstack.org/#/c/378930/ | 13:32 |
| odyssey4me | evrardjp yes | 13:33 |
| odyssey4me | my primary motivation would be that service restarts no longer get failures ignored | 13:33 |
| drifterza | ah yes | 13:33 |
| drifterza | evrardjp, thats what I meant | 13:33 |
| *** galstrom_zzz is now known as galstrom | 13:34 | |
| odyssey4me | evrardjp I know it can be improved upon like we discussed - but based on the existing implementations I think that is good enough as a small evolution. | 13:34 |
| evrardjp | I know the motivation, and surprisingly I can read :D | 13:34 |
| evrardjp | I have no strong opinion, and I'd be happy to leave the community decide | 13:35 |
| evrardjp | It's a late "feature" for inclusion | 13:35 |
| odyssey4me | sure, it is | 13:35 |
| evrardjp | but osa has always been a different beast, I don't really see the problem | 13:35 |
| evrardjp | so... grey! | 13:36 |
| evrardjp | or transparent | 13:36 |
| evrardjp | whatever | 13:36 |
| odyssey4me | lemme propose the changes and we can duke it out in reviews | 13:36 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder: Revert "Add retries to upstart/systemd reload" https://review.openstack.org/386616 | 13:37 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder: Revert "Add retries to upstart/systemd reload" https://review.openstack.org/386616 | 13:38 |
| *** jmckind has joined #openstack-ansible | 13:38 | |
| *** Andrew_jedi_ has joined #openstack-ansible | 13:40 | |
| *** Andrew_jedi has quit IRC | 13:42 | |
| *** Andrew_jedi_ is now known as Andrew_jedi | 13:42 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat: Use dictionary for service group mappings https://review.openstack.org/386621 | 13:43 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh: Use dictionary for service group mappings https://review.openstack.org/380162 | 13:50 |
| *** admin0_ has joined #openstack-ansible | 13:50 | |
| *** pester has joined #openstack-ansible | 13:51 | |
| *** admin0 has quit IRC | 13:53 | |
| *** admin0_ is now known as admin0 | 13:53 | |
| *** fxpester has quit IRC | 13:53 | |
| *** techstepper has joined #openstack-ansible | 13:59 | |
| *** sdake has joined #openstack-ansible | 14:00 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder: Use dictionary for service group mappings https://review.openstack.org/380165 | 14:01 |
| *** sdake_ has quit IRC | 14:04 | |
| openstackgerrit | Frédéric Guillot proposed openstack/openstack-ansible-os_almanach: [WIP] Make project compatible with CI infra https://review.openstack.org/386233 | 14:06 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: Use dictionary for service group mappings https://review.openstack.org/380164 | 14:07 |
| *** Mudpuppy has joined #openstack-ansible | 14:09 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova: Use dictionary for service group mappings https://review.openstack.org/386630 | 14:11 |
| andymccr | odyssey4me: that magnum testing - it seems like tempest itself takes 20 mins on successful runs too. | 14:11 |
| *** Andrew_jedi has quit IRC | 14:11 | |
| odyssey4me | andymccr perhaps we should choose a better test then? | 14:11 |
| odyssey4me | something good enough, but shorter | 14:11 |
| andymccr | odyssey4me: yeah i think that is probably the best approach | 14:11 |
| andymccr | although saying that - i do think we have an issue with xenial tests | 14:12 |
| andymccr | they seem to take between 10-20 mins longer than trusty on a LOT of gates | 14:12 |
| *** todd_m_dav has joined #openstack-ansible | 14:14 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift: Use dictionary for service mappings https://review.openstack.org/386635 | 14:17 |
| logan- | andymccr: agreed.. i haven't dug into this yet but my local CI results are like a 30-40min increase. once we get trusty fixed up i'll be interested to see if it is newton or xenial behind the huge slowdown | 14:17 |
| logan- | http://cdn.pasteraw.com/l3t17rmafnaug09ddhce68nj3k2phx6 | 14:17 |
| odyssey4me | hmm, I would have thought Xenial would be faster | 14:18 |
| odyssey4me | logan- do you have any CI testing going for newton+trusty? | 14:19 |
| logan- | not yet due to the cow container issue | 14:19 |
| odyssey4me | I can understand newton being slower, simply due to all the additional code for trusty vs xenial vs centos (systemd vs upstart) | 14:19 |
| logan- | once the overlay fix is backported to newton I should have newton+trusty ci going | 14:19 |
| odyssey4me | also mitaka was Ansible 1.9 while newton is 2.1.1 - so I'd like to see just one variable change - trusty vs xenial for newton | 14:20 |
| odyssey4me | logan- that it alsready backported | 14:20 |
| logan- | oh | 14:20 |
| odyssey4me | *already | 14:20 |
| logan- | well then | 14:20 |
| logan- | let's find out :) | 14:20 |
| *** allanice001 has quit IRC | 14:20 | |
| odyssey4me | it merged last night | 14:20 |
| *** spotz_zzz is now known as spotz | 14:21 | |
| todd_m_dav | Hi all. I'm looking for some help. I am having an issue trying to install OSA 13.3.4. In the setup-hosts.yml section, I get failures configuring the volume containers. "fatal: [myhost-04_cinder_volumes_container-2c17476d -> myhost-04] => One or more undefined variables: 'dict object' has no attribute 'interface' | 14:22 |
| openstackgerrit | Frédéric Guillot proposed openstack/openstack-ansible-os_almanach: [WIP] Make project compatible with CI infra https://review.openstack.org/386233 | 14:22 |
| logan- | thanks odyssey4me. got some jobs queued up now. will check back in a while | 14:22 |
| todd_m_dav | It looks to me like facts are not getting generated for those containers, but I am a bit stuck figuring out why. | 14:23 |
| todd_m_dav | If that is even the issue... | 14:23 |
| *** chris_hultin|AWA is now known as chris_hultin | 14:23 | |
| odyssey4me | todd_m_dav you can try clearing facts, which may need to be done if you've been fiddling with the config in the last 24 hours | 14:25 |
| odyssey4me | otherwise you may have an issue with your config | 14:26 |
| *** berendt has quit IRC | 14:26 | |
| odyssey4me | http://docs.openstack.org/developer/openstack-ansible/mitaka/install-guide/ops-troubleshooting.html#cached-ansible-facts-issues | 14:26 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: [WIP] Attempt to fix swift_pypy memory issues https://review.openstack.org/386566 | 14:31 |
| todd_m_dav | oddesy4me: I did clear that directory from the deploy host before starting the openstack-ansible run, and deleted the 3 cinder_volume container fact files. Files are created for those containers, but they don't get populated with any JSON. It's possible that it is a parsing error but I can't find anything wrong with the openstack_inventory.json | 14:32 |
| todd_m_dav | Thanks for the link. I will see if I can find any debugging tips there | 14:32 |
| *** fxpester has joined #openstack-ansible | 14:33 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests: Implement centralised Ansible test scripts https://review.openstack.org/386642 | 14:33 |
| odyssey4me | todd_m_dav it sounds suspiciously like you have a config error then - either that or the inventory.json has a previously populated set of data that's broken | 14:34 |
| andymccr | logan-: on your time runs are you deploying swift? | 14:34 |
| odyssey4me | we've made the inventory a little more 'self healing' in newton | 14:34 |
| odyssey4me | todd_m_dav check the indentations for your yaml very carefully | 14:34 |
| *** pester has quit IRC | 14:35 | |
| *** woodard has quit IRC | 14:37 | |
| todd_m_dav | oddesy4me, yeah, that sounds likely. I'll go back to the well and see if I can't find something off. So much for the hope of hitting a know issue :D | 14:37 |
| todd_m_dav | Thanks much | 14:37 |
| *** sdake_ has joined #openstack-ansible | 14:38 | |
| *** sdake_ has quit IRC | 14:40 | |
| *** vishwanathj has joined #openstack-ansible | 14:41 | |
| *** sdake has quit IRC | 14:41 | |
| cloudnull | andymccr: you about? | 14:42 |
| cloudnull | https://bugs.launchpad.net/openstack-ansible/+bug/1633221 -- is this something that should be worked on ? | 14:42 |
| openstack | Launchpad bug 1633221 in openstack-ansible "OSA creates reseller_admin role but role should be ResellerAdmin" [Undecided,New] - Assigned to Praveen N (praveenn) | 14:42 |
| admin0 | evrardjp: it did not happen .. the ips now came up instantly ( same hardware, same setup where it failed multiple times before ) .. the only change is latest checkouts and new bootstrap-ansible | 14:43 |
| *** pester has joined #openstack-ansible | 14:44 | |
| *** michauds has joined #openstack-ansible | 14:44 | |
| andymccr | cloudnull: looking | 14:44 |
| *** fxpester has quit IRC | 14:44 | |
| *** pester has quit IRC | 14:45 | |
| andymccr | cloudnull: hmm thats weird we set it correctly from what I can see... | 14:46 |
| andymccr | maybe in older version of OSA - can take a look but it seems correct in master | 14:46 |
| andymccr | swift_reselleradmin_role: ResellerAdmin | 14:46 |
| *** karimb has quit IRC | 14:46 | |
| odyssey4me | maybe liberty or even kilo | 14:47 |
| odyssey4me | I'm fairly sure we implemented a fix for that some time in the newton cycle and probably backported it | 14:47 |
| *** cmart has joined #openstack-ansible | 14:47 | |
| odyssey4me | although it could have been in the mitaka cycle - who knows | 14:47 |
| andymccr | oh we're only creating the ResellerAdmin role if you're using ceilometer | 14:49 |
| andymccr | that might be a bug | 14:49 |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-os_nova: Add CentOS support for os_nova https://review.openstack.org/378963 | 14:49 |
| *** karimb has joined #openstack-ansible | 14:49 | |
| *** janki has quit IRC | 14:49 | |
| logan- | odyssey4me: i believe so. it uses gate-check-commit | 14:49 |
| *** haad2 has joined #openstack-ansible | 14:49 | |
| *** haad1 has quit IRC | 14:50 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Remove the rabbitmq deterministic sort https://review.openstack.org/386650 | 14:51 |
| admin0 | someone said we will not see too many 503’s .. i see too many of them :D | 14:52 |
| *** sdake has joined #openstack-ansible | 14:53 | |
| odyssey4me | logan- I think you meant that for andymccr - ie it is executing swift because it's doing the same job as we do in the integrated build | 14:53 |
| logan- | oh yes sorry. yes it is running gate-check-commit, so unless something has changed with newton I think it runs swift | 14:54 |
| odyssey4me | andymccr thanks to mgariepy we're able to promote centos7 jobs to voting for os_neutron: https://review.openstack.org/386651 | 14:54 |
| logan- | @ andymccr | 14:54 |
| *** weezS has joined #openstack-ansible | 14:54 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: Always setup ResellerAdmin role in keystone https://review.openstack.org/386652 | 14:54 |
| andymccr | logan-: ok cool - i think there is an issue with the "service: x, state: started" job in swift - which is odd because thats really straight forward (or should be) | 14:55 |
| andymccr | but takes a long time in Swift for some reason, on xenial. | 14:55 |
| logan- | ahh interesting | 14:56 |
| logan- | TASK: os_swift : Ensure services are started -------------------------- 566.77s | 14:56 |
| logan- | heh | 14:56 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: Always setup ResellerAdmin role in keystone https://review.openstack.org/386652 | 14:58 |
| mgariepy | yay ! | 14:58 |
| mgariepy | nova on centos is... painful. | 14:58 |
| andymccr | logan-: yeah its weird, that task jsut does a "state: started" | 15:00 |
| andymccr | so i dont see why it takes 10 mins on xenial and 30s on trusty... | 15:01 |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 15:01 |
| logan- | well my xenial job is at keystone right now, so let me see if i can catch it during swift and look at what's spinning during that 10 minute window | 15:02 |
| drifterza | hi andymccr | 15:02 |
| odyssey4me | stevelle does the ceilometer role perhaps also need something like https://review.openstack.org/#/c/384002/1/defaults/main.yml ? I see the API services failing, but also that it appears that ceilometer has Apache serving something | 15:02 |
| *** agrebennikov has joined #openstack-ansible | 15:02 | |
| andymccr | logan-: awesome, im trying a few things that are more work arounds than fixers because it seems like an issue with the module itself. | 15:02 |
| andymccr | hey drifterza how goes? | 15:02 |
| drifterza | good man good, you ? | 15:03 |
| andymccr | yeah all good | 15:03 |
| *** sdake_ has joined #openstack-ansible | 15:03 | |
| drifterza | how suck am I 98 patchsets for 1 role | 15:04 |
| drifterza | I must be a masecast | 15:04 |
| andymccr | drifterza: hahaha - its kinda hard to imp a whole role in one go | 15:05 |
| drifterza | yeah dude, I'll never do it again | 15:05 |
| andymccr | its probably easiest to have like a convergence test to start | 15:05 |
| *** sdake has quit IRC | 15:05 | |
| andymccr | and then slowly imp the rest of it | 15:05 |
| andymccr | are you far off? | 15:06 |
| cloudnull | mgariepy: idk if this is helpful but i had a fork from the kilo time which did our roles on centos https://github.com/cloudnull/os-ansible-deployment/tree/master/playbooks/roles/os_nova | 15:06 |
| drifterza | nah since ive been including external roles for dependancies its going better | 15:06 |
| drifterza | so close | 15:06 |
| cloudnull | i remember there being plenty of issues, the main one i had to overcome was pollicy kit | 15:06 |
| drifterza | well I hope its close lol | 15:06 |
| andymccr | excellent - happy to take a look if you hit blockers. | 15:07 |
| drifterza | excellent thanks man, appreciate it :) | 15:07 |
| cloudnull | but there may be things in the forked role that could be of help, maybe. | 15:07 |
| mgariepy | cloudnull, anything about a pre-historic version of qemu ? | 15:07 |
| mgariepy | cloudnull, http://logs.openstack.org/63/378963/33/check/gate-openstack-ansible-os_nova-ansible-func-centos-7-nv/92e1324/logs/host/nova/nova-compute.log.txt.gz#_2016-10-14_14_03_33_792 | 15:08 |
| cloudnull | mgariepy: i dont remember anything with dated qemu | 15:09 |
| drifterza | cloudnull, I suggested the IPv6 provider stuff to our networks team, they laughed at me saying is ipv6 still a thing, its been 10yrs and still no major adoption | 15:09 |
| drifterza | the sigh | 15:09 |
| cloudnull | le sign | 15:09 |
| drifterza | but they gonna give me a /32 | 15:09 |
| cloudnull | reply "nat is a hack" | 15:09 |
| cloudnull | :p | 15:09 |
| drifterza | its like a trillion ips | 15:10 |
| evrardjp | or just one | 15:10 |
| drifterza | ipv6/32 | 15:10 |
| evrardjp | you've been tricked into thinking it was v6! | 15:10 |
| drifterza | i mean | 15:10 |
| cloudnull | cloud1 is a dual stack 10/8 and v6/64 | 15:10 |
| drifterza | hahah evrardjp | 15:10 |
| cloudnull | so we're limited to around 16 million ips | 15:10 |
| cloudnull | it'll be good enough for now. :) | 15:11 |
| drifterza | im going to do ipv6 and when they come around ill be, yeah boooy its done. | 15:11 |
| cloudnull | that's my approach. | 15:11 |
| odyssey4me | mgariepy if it's not already happening, CentOS should pretty much have RDO setup everywhere. It should not have to use EPEL either - I believe that the RDO crew have ensured that they have all the packages needed for an OpenStack environment. | 15:11 |
| drifterza | yeah, people are so short sighted | 15:11 |
| evrardjp | v4? so last century | 15:11 |
| odyssey4me | mgariepy that may give you better versions of qemu, libvirt, etc | 15:11 |
| evrardjp | so last millenium | 15:11 |
| cloudnull | let the corp spin up a pm team and deepdive then just hand over the work and say, you're welcome | 15:11 |
| drifterza | lol cloudnull, sad part is, they get paid the big cash | 15:12 |
| cloudnull | yea. that's a thing. | 15:12 |
| odyssey4me | drifterza you should inform them that they are mistaken - RAX, for one as a provider, has IPv6 in its pub cloud | 15:13 |
| drifterza | I need to convince rackspace that africa is not in mexico and they should get a presence here in za. | 15:13 |
| cloudnull | hahahaha | 15:13 |
| drifterza | so i can work for a decent place | 15:13 |
| cloudnull | jcannava: ^ | 15:13 |
| admin0 | drifterza: good deal :) | 15:13 |
| cloudnull | i literally just LOL'd | 15:13 |
| odyssey4me | Rackspace knows very well where Africa is - manh of its staff come from South African. :) | 15:13 |
| odyssey4me | *many | 15:14 |
| odyssey4me | *South Africa | 15:14 |
| odyssey4me | bah, friday | 15:14 |
| admin0 | mexico of africa :D | 15:14 |
| drifterza | odyssey4me, they should do something here. | 15:14 |
| drifterza | so much talent | 15:14 |
| drifterza | and I don't have to move to the uk then | 15:14 |
| drifterza | :P | 15:14 |
| admin0 | drifterza: talk about a franchaise with them | 15:14 |
| admin0 | start your own | 15:14 |
| admin0 | solved \o/ | 15:14 |
| drifterza | lol | 15:14 |
| cloudnull | we can franchise and federate. its just that easy. | 15:14 |
| cloudnull | sorry ^ old joke from the portland OS summit. | 15:15 |
| drifterza | haha | 15:15 |
| cloudnull | i laughed then i relized i'm the only one. | 15:15 |
| *** pcaruana has quit IRC | 15:15 | |
| admin0 | i am off to home .. plan to play x-plane 10 with a flightsim yoke and a full screen HD projector :D | 15:15 |
| cloudnull | nice! | 15:15 |
| drifterza | niiice | 15:15 |
| admin0 | and some league of legends :) | 15:15 |
| admin0 | and cities skylines | 15:16 |
| cloudnull | you should play rocket league | 15:16 |
| cloudnull | that game is mind numbing fun | 15:16 |
| admin0 | but i do plan to figure out this multi region stuff this weekend as well | 15:16 |
| drifterza | cloudnull, dude watching jaws 1 atm, I'm so nostalgic atm | 15:16 |
| cloudnull | I love that movie. | 15:16 |
| *** Oku_OS is now known as Oku_OS-away | 15:16 | |
| drifterza | classic | 15:17 |
| cloudnull | i watch old movies all the time. my wife hates it :) | 15:17 |
| drifterza | yeah mine too | 15:17 |
| drifterza | if its not twilight and me trying to joke no my own puke | 15:17 |
| drifterza | choke* | 15:17 |
| drifterza | lol | 15:17 |
| cloudnull | i made her suffer through the rocky trillogy a few months back . | 15:18 |
| admin0 | i might soon do something like that: https://cdn-images-1.medium.com/max/800/1*k31CzsZOtqA89PBVNN5zWA.jpeg — but with split screen and multiple osa runs :D | 15:18 |
| odyssey4me | andymccr release requests for liberty/mitaka: https://review.openstack.org/386585 / https://review.openstack.org/386569 | 15:18 |
| cloudnull | it was epic . :) | 15:18 |
| drifterza | dude my wife loves rocky | 15:18 |
| drifterza | she wants to box me, everytime she's done watching it | 15:18 |
| cloudnull | admin0: that'd be amazing. | 15:18 |
| cloudnull | we watched them all to get ready to see creed . | 15:19 |
| drifterza | have you seen the latest rocky, damn sylvester looks horrible | 15:19 |
| cloudnull | which sh loved. | 15:19 |
| drifterza | looks like jason x | 15:19 |
| cloudnull | hahaha | 15:19 |
| admin0 | i am out .. have a great time all | 15:19 |
| cloudnull | yea. he's getting ol.d. | 15:19 |
| drifterza | later admin0 | 15:19 |
| cloudnull | have a good one admin0 | 15:19 |
| admin0 | drifterza: i am redoing the env .. will do multi region this weekend .. keys will remain :) | 15:19 |
| odyssey4me | so admin0 you mean like http://esreality.com/files/inlineimages/2015/104254-102%20bolus.jpg ? | 15:19 |
| cloudnull | raging bull is another good one. | 15:19 |
| admin0 | :D | 15:19 |
| drifterza | nice | 15:19 |
| drifterza | me=rambo +2 | 15:20 |
| drifterza | love those | 15:20 |
| *** admin0 has quit IRC | 15:20 | |
| cloudnull | oh. yea. classics. | 15:20 |
| drifterza | even the last one, brutal when he mows down those vietnamese okes with that gun | 15:20 |
| drifterza | stupid missionaries and he has to go save them. | 15:21 |
| drifterza | 1 guy vs an army | 15:21 |
| odyssey4me | Predator :) | 15:21 |
| mgariepy | odyssey4me, https://www.redhat.com/archives/rdo-list/2016-August/msg00210.html | 15:21 |
| cloudnull | another classic. | 15:21 |
| openstackgerrit | Frédéric Guillot proposed openstack/openstack-ansible-os_almanach: [WIP] Make project compatible with CI infra https://review.openstack.org/386233 | 15:21 |
| drifterza | oooh yeah | 15:21 |
| drifterza | damn | 15:21 |
| cloudnull | we're doing 5'th element this weekend | 15:21 |
| drifterza | LOL | 15:21 |
| drifterza | classic | 15:22 |
| cloudnull | she's never seen it so we have to correct the situation . | 15:22 |
| drifterza | OH | 15:22 |
| drifterza | damn | 15:22 |
| odyssey4me | oh wow, awesome :) | 15:22 |
| drifterza | you see theres a new lethal weapon series out | 15:22 |
| cloudnull | lolol. | 15:22 |
| cloudnull | no. | 15:22 |
| drifterza | yeah, not the movie though | 15:22 |
| cloudnull | interesting. | 15:22 |
| drifterza | with one of the wayans bro's | 15:22 |
| * cloudnull goes to see. | 15:22 | |
| drifterza | it seems okay, I dig the wayans bro's tho | 15:23 |
| cloudnull | is it any good? | 15:23 |
| drifterza | its just another cop show, not anything like blue bloods | 15:23 |
| drifterza | damn I dig Tom Selek | 15:23 |
| *** michaelgugino has joined #openstack-ansible | 15:23 | |
| drifterza | the Magnum PI, he's mostashe is legendary | 15:24 |
| cloudnull | indeed. | 15:24 |
| *** chhavi has joined #openstack-ansible | 15:25 | |
| drifterza | awesome jaws gonna eat someone | 15:25 |
| cloudnull | lethal weapon added to the queue. i'll check it out. | 15:25 |
| drifterza | go boy go | 15:25 |
| drifterza | leg off | 15:25 |
| drifterza | nice | 15:25 |
| drifterza | anyone else sick of america's got talent and X factor | 15:26 |
| drifterza | can they stop with that sh* now | 15:26 |
| cloudnull | the wife likes the voice | 15:27 |
| cloudnull | i'm not a fan of contestant shows. | 15:27 |
| drifterza | jeeze I saw that the other day, christina looks like shit | 15:27 |
| drifterza | the makeup is like draggg almost | 15:28 |
| cloudnull | hahaha. | 15:28 |
| cloudnull | i say the same things. | 15:28 |
| cloudnull | :) | 15:28 |
| drifterza | ;) | 15:29 |
| drifterza | apricot brandy, seriously | 15:29 |
| drifterza | never heard of it | 15:29 |
| drifterza | jeeze I owned myself today, we rented one of those massive jumping castles that goes into the pool, that thing is heavy and hard to setup. | 15:30 |
| cloudnull | yikes. | 15:31 |
| drifterza | its one of those that you put the hose in and makes it slippery for the kids | 15:32 |
| cloudnull | sounds like a blast. | 15:32 |
| drifterza | I must say your dogs are cute | 15:32 |
| cloudnull | hahaha. | 15:32 |
| cloudnull | you see the last pics :) | 15:33 |
| drifterza | yeah | 15:33 |
| drifterza | i loled | 15:33 |
| cloudnull | https://twitter.com/cloudnull/status/786740834347601920 | 15:33 |
| cloudnull | hahahaha | 15:33 |
| *** allanice001 has joined #openstack-ansible | 15:33 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-os_nova: Add CentOS support for os_nova https://review.openstack.org/378963 | 15:33 |
| evrardjp | hehe | 15:33 |
| drifterza | how old ? | 15:33 |
| cloudnull | the white one is about 8 mo, the tan one is about 18 mo. | 15:34 |
| drifterza | nice | 15:35 |
| *** javeriak has joined #openstack-ansible | 15:36 | |
| drifterza | oh yeah thats right, halloween is in a few weeks | 15:36 |
| *** johnmilton has quit IRC | 15:36 | |
| drifterza | we don't celebrate it | 15:36 |
| *** johnmilton has joined #openstack-ansible | 15:36 | |
| *** johnmilton has joined #openstack-ansible | 15:37 | |
| *** adrian_otto has joined #openstack-ansible | 15:40 | |
| allanice001 | drifterza: we don’t celerate haloween, we just perpetually look like zomies @ work :P | 15:45 |
| drifterza | haha | 15:45 |
| drifterza | thought you were on leave dude ? | 15:46 |
| allanice001 | only from monday | 15:46 |
| allanice001 | still got the laptop out, trying to wrap up few stuff | 15:46 |
| drifterza | ah okay | 15:47 |
| openstackgerrit | Frédéric Guillot proposed openstack/openstack-ansible-os_almanach: [WIP] Make project compatible with CI infra https://review.openstack.org/386233 | 15:48 |
| *** batkins61 has quit IRC | 15:49 | |
| cloudnull | any chance cores will have a moment to have a look at getting https://review.openstack.org/#/c/384635/ though ? | 15:53 |
| fguillot | what is the best way to run functional tests for a role on my local machine before to send a new review? actually I'm using vagrant with xenial64, I would like to replicate the same setup as the jenkins slave in local | 15:54 |
| *** sdake has joined #openstack-ansible | 15:54 | |
| logan- | hey andymccr: http://cdn.pasteraw.com/10fkc08y6xd42iu943fh2i19cuoswvu | 15:54 |
| logan- | the reason it takes 10 minutes is because the unit file keeps retrying to start it | 15:55 |
| *** sdake_ has quit IRC | 15:56 | |
| allanice001 | fguillot: is your role standalone? | 15:57 |
| logan- | im guessing the reason that service task doesnt do it on trusty is because upstart fails it immediately instead of retrying the restart for 10 minutes in the background | 15:57 |
| fguillot | allanice001 I don't have any dependencies at the moment, see https://review.openstack.org/#/c/386233 | 15:58 |
| logan- | andymccr: same thing with swift-object-replicator-server | 15:59 |
| allanice001 | fguillot: you’re welcome to take a squizz at https://github.com/allanice001/ansible-base | 15:59 |
| allanice001 | the galaxy role is available as allanice001.ansible-base | 15:59 |
| allanice001 | i got a vagrant file there | 16:00 |
| allanice001 | i use this setup / structure to test with | 16:00 |
| andymccr | logan-: it looks like its the replicator services | 16:01 |
| logan- | yeah, those bins that systemd is trying to execute don't exist in the venv | 16:01 |
| fguillot | allanice001 ok, I will take a look | 16:01 |
| logan- | i locked the jenkins node so it won't delete. let me know if you want to ssh in and dig around the aio | 16:02 |
| *** hughmFLEXin has quit IRC | 16:04 | |
| *** dxiri has joined #openstack-ansible | 16:04 | |
| *** javeriak has quit IRC | 16:05 | |
| odyssey4me | cloudnull waiting for mattt and evrardjp to test that as they're still busy with work on that front | 16:05 |
| evrardjp | I'm reading it right now | 16:05 |
| odyssey4me | fguillot if you're using vagrant - see our other roles for a template | 16:06 |
| evrardjp | it doesn't sound an heresy, it will just make us rework the docs :p | 16:06 |
| evrardjp | alextricity25: also worked on this | 16:06 |
| odyssey4me | fguillot you'll need to add a vagrantfile, and run_tests.sh | 16:06 |
| odyssey4me | fguillot the vangrantfile sets the parameters for the VM, then run_tests does some minimal setup before going ahead and executing tox just like the gate tests do | 16:06 |
| odyssey4me | fguillot none of our other roles use https://github.com/allanice001/ansible-base - if that's considered, then ideally it should evolve from what we already have | 16:07 |
| fguillot | odyssey4me looking at gnocchi project, the vagrantfile uses the image trusty64 and I have a gate for xenial, that won't make any difference? | 16:08 |
| odyssey4me | fguillot sure, just change that to ubuntu/xenial64 | 16:08 |
| allanice001 | odyssey4me: i was pointing to it for the vagrant file which cosumes the role - for local testing | 16:09 |
| odyssey4me | some of the roles are a little behind on the updates :) | 16:09 |
| fguillot | ok | 16:09 |
| odyssey4me | fguillot for example https://github.com/openstack/openstack-ansible-rabbitmq_server/blob/master/Vagrantfile | 16:09 |
| evrardjp | odyssey4me: sometimes I think we should have a boilerplate role that installs stuff and we just throw variables | 16:09 |
| odyssey4me | but we'll probably pull out all the trusty stuff in the Ocata cycle | 16:09 |
| *** aludwar has quit IRC | 16:10 | |
| *** karimb has quit IRC | 16:10 | |
| evrardjp | then I come back to my senses, and realize ansible dependencies sometimes suck (with variables), include_roles isn't great, and self-contained is a nice thing to have | 16:10 |
| evrardjp | but sometimes I dream of a perfect ansible and DRY code | 16:11 |
| *** haad2 has quit IRC | 16:11 | |
| *** javeriak has joined #openstack-ansible | 16:11 | |
| allanice001 | evrardjp: if you maintain a proper requirements file, you could just run ansible-galaxy | 16:12 |
| *** hughmFLEXin has joined #openstack-ansible | 16:12 | |
| allanice001 | even if role is pulled from git | 16:12 |
| *** haad1 has joined #openstack-ansible | 16:13 | |
| odyssey4me | allanice001 we already do that | 16:13 |
| mgariepy | anyone have a script to download recursivly the logs of the gate checks ? | 16:13 |
| odyssey4me | logan- aha - I see cc andymccr - you know, perhaps when we lay down the upstart/systemd scripts we should have them disabled - then have the handler remove the disable when it executes the restart? | 16:14 |
| allanice001 | yeah, i know odyssey4me , but in our environment, we may have private brances of a specific component we want to test internally | 16:14 |
| odyssey4me | that way we control exactly when the service starts, stops and restarts | 16:14 |
| odyssey4me | allanice001 which is why we allow pretty much every git source we use to be overrideen | 16:15 |
| odyssey4me | *overridden | 16:15 |
| *** cmart has quit IRC | 16:15 | |
| evrardjp | allanice001: I was thinking further, like one role for systemd configuration and you just give a few variables, etc... but you know that's just dreaming :p | 16:16 |
| allanice001 | some people have done some stuff in that arena | 16:16 |
| allanice001 | its slightly dated, but take a squiz through the debops roles | 16:17 |
| allanice001 | github.com/debops | 16:17 |
| *** asettle__ has joined #openstack-ansible | 16:17 | |
| evrardjp | I know :) | 16:17 |
| logan- | odyssey4me: i'm not familiar with swift, but just looking at the venv, how would those restarts ever work? these bins swift-object-replicator-server and swift-container-replicator-server aren't in the venv at all | 16:18 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: Fix swift init scripts w/o dedicated replication https://review.openstack.org/386679 | 16:18 |
| andymccr | odyssey4me: logan- ^ | 16:18 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Automatically source the rc file for all Ansible commands https://review.openstack.org/386680 | 16:19 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: Fix swift init scripts w/o dedicated replication https://review.openstack.org/386679 | 16:19 |
| logan- | thanks andymccr | 16:20 |
| odyssey4me | andymccr maybe use 'service_en' instead? https://github.com/openstack/openstack-ansible-os_neutron/blob/master/defaults/main.yml#L200 | 16:20 |
| *** asettle has quit IRC | 16:21 | |
| *** aludwar has joined #openstack-ansible | 16:21 | |
| odyssey4me | andymccr otherwise we're also using https://github.com/openstack/openstack-ansible-os_cinder/blob/master/defaults/main.yml#L260 | 16:21 |
| *** asettle__ has quit IRC | 16:21 | |
| odyssey4me | andymccr with https://github.com/openstack/openstack-ansible-os_cinder/blob/master/handlers/main.yml#L43 | 16:22 |
| odyssey4me | and https://github.com/openstack/openstack-ansible-os_cinder/blob/master/tasks/cinder_init_common.yml#L30 | 16:24 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: Fix swift init scripts w/o dedicated replication https://review.openstack.org/386679 | 16:25 |
| andymccr | fixed | 16:25 |
| *** xek_ has quit IRC | 16:25 | |
| andymccr | or now fixed :P | 16:26 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: Fix swift init scripts w/o dedicated replication https://review.openstack.org/386679 | 16:26 |
| *** xek_ has joined #openstack-ansible | 16:26 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_magnum: Only run cluster tests https://review.openstack.org/386686 | 16:27 |
| *** allanice001 has quit IRC | 16:35 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Automatically source the rc file for all Ansible commands https://review.openstack.org/386680 | 16:35 |
| *** sdake_ has joined #openstack-ansible | 16:38 | |
| *** adrian_otto has quit IRC | 16:39 | |
| *** adrian_otto has joined #openstack-ansible | 16:40 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Automatically source the rc file for all Ansible commands https://review.openstack.org/386680 | 16:40 |
| *** sdake has quit IRC | 16:40 | |
| *** chhavi has quit IRC | 16:41 | |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 16:41 |
| evrardjp | cloudnull: doing the testing in a bit, first have to redeploy liberty | 16:42 |
| evrardjp | but afk | 16:42 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Automatically source the rc file for all Ansible commands https://review.openstack.org/386680 | 16:42 |
| *** adrian_otto has quit IRC | 16:43 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Automatically source the rc file for all Ansible commands https://review.openstack.org/386680 | 16:52 |
| *** xek_ has quit IRC | 16:55 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-lxc_container_create: Drop container setup script independent of container_networks https://review.openstack.org/386699 | 16:56 |
| *** xek_ has joined #openstack-ansible | 16:56 | |
| odyssey4me | ok, I'm out for the weekend - have a great w/end all! | 16:59 |
| drifterza | cheers dude | 16:59 |
| *** phschwartz has quit IRC | 17:00 | |
| *** chris_hultin is now known as chris_hultin|AWA | 17:01 | |
| *** techstepper has quit IRC | 17:06 | |
| *** adrian_otto has joined #openstack-ansible | 17:08 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-lxc_container_create: Don't fail veth-cleanup template when no container_networks https://review.openstack.org/386700 | 17:10 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: Always setup ResellerAdmin role in keystone https://review.openstack.org/386652 | 17:11 |
| *** michauds has quit IRC | 17:12 | |
| stevelle | odyssey4me: I'm not sure what you asked (2h ago) | 17:13 |
| *** woodard has joined #openstack-ansible | 17:14 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_magnum: Only run cluster tests https://review.openstack.org/386686 | 17:16 |
| *** omiday has quit IRC | 17:17 | |
| *** omiday has joined #openstack-ansible | 17:17 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: Fix swift init scripts w/o dedicated replication https://review.openstack.org/386679 | 17:19 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: Fix swift init scripts w/o dedicated replication https://review.openstack.org/386679 | 17:20 |
| andymccr | odyssey4me: logan-: one more fix ^ noticed the normal gate was still taking ages - the program_binary option wasnt being used for systemd. that should fix that one. | 17:20 |
| *** haad1 has quit IRC | 17:21 | |
| *** allanice001 has joined #openstack-ansible | 17:27 | |
| odyssey4me | stevelle os_ceilometer is failing for both master and newton - I added some comments in https://review.openstack.org/380164 | 17:31 |
| odyssey4me | basically I suspect that both Apache and upstart/systemd might be trying to run the same service | 17:32 |
| stevelle | odyssey4me: I can look at updating https://review.openstack.org/#/c/385697/ to remove some stuff | 17:33 |
| stevelle | but look at the failure there | 17:33 |
| odyssey4me | stevelle cool - appreciate you looking into it | 17:34 |
| stevelle | I haven't had time to look at the tempest failure | 17:34 |
| odyssey4me | wow, that's a very lengthy output - difficult to work through :/ | 17:36 |
| *** phschwartz has joined #openstack-ansible | 17:40 | |
| *** karimb has joined #openstack-ansible | 17:42 | |
| stevelle | ikr | 17:43 |
| *** rgogunskiy has joined #openstack-ansible | 17:44 | |
| *** rgogunskiy has quit IRC | 17:44 | |
| *** cmart has joined #openstack-ansible | 17:46 | |
| openstackgerrit | Merged openstack/openstack-ansible-haproxy_server: Fix markup error https://review.openstack.org/386562 | 17:46 |
| *** javeriak has quit IRC | 17:47 | |
| stevelle | I'll go through ceilo and make sure we clean up all the rest of the api service cruft at the same time in 385697 but that doesn't immediately appear to be what is causing the tempest failure | 17:47 |
| *** pcaruana has joined #openstack-ansible | 17:48 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-haproxy_server: Fix markup error https://review.openstack.org/386715 | 17:49 |
| odyssey4me | cloudnull I think you missed the point entirely in https://review.openstack.org/386680 | 17:50 |
| cloudnull | i dont think so . | 17:51 |
| cloudnull | but please explain | 17:51 |
| odyssey4me | brb | 17:51 |
| odyssey4me | see reponses in review | 17:51 |
| *** javeriak has joined #openstack-ansible | 17:51 | |
| *** javeriak has quit IRC | 17:52 | |
| *** Jeffrey4l_ has quit IRC | 17:52 | |
| cloudnull | that commit forces folks to use our rc | 17:53 |
| cloudnull | which breaks everything else not openstack-ansible | 17:53 |
| cloudnull | or could i should say | 17:54 |
| cloudnull | it ties it to our system which we should not do | 17:54 |
| cloudnull | i should be able to run an ansible command without interacting with any of the osa machinery | 17:56 |
| openstackgerrit | Merged openstack/openstack-ansible-haproxy_server: Fix markup error https://review.openstack.org/386715 | 17:56 |
| *** cmart has quit IRC | 17:58 | |
| *** thorst_ has quit IRC | 18:03 | |
| odyssey4me | cloudnull well, it's easy enough to make the wrapper only source the rc file if you're executing from the right place | 18:04 |
| odyssey4me | ie from somewhere in /opt/openstack-ansible | 18:04 |
| *** thorst_ has joined #openstack-ansible | 18:04 | |
| *** jcrst has joined #openstack-ansible | 18:04 | |
| cloudnull | in some cases though a deployer may not be using /opt/openstack-ansible | 18:05 |
| odyssey4me | sure, no worries - will figure that part out | 18:05 |
| odyssey4me | I don't like the forced wrapper much either - but I also don't like the fact that we keep getting people stuck. | 18:06 |
| odyssey4me | so I'm just trying to find the middle ground | 18:06 |
| *** chris_hultin|AWA is now known as chris_hultin | 18:06 | |
| cloudnull | that would create osa-* wrappers [ http://cdn.pasteraw.com/tmobe5n62ll2ebjw0x2jnoxan5dbol9 ] then native commands would work without any tomfoolery. | 18:07 |
| odyssey4me | bah, I really don't like that idea either | 18:08 |
| *** javeriak has joined #openstack-ansible | 18:08 | |
| openstackgerrit | Frédéric Guillot proposed openstack/openstack-ansible-os_almanach: [WIP] Make project compatible with CI infra https://review.openstack.org/386233 | 18:10 |
| *** gouthamr_ has joined #openstack-ansible | 18:11 | |
| *** thorst_ has quit IRC | 18:12 | |
| *** gouthamr has quit IRC | 18:13 | |
| *** fops has joined #openstack-ansible | 18:14 | |
| *** pabelanger has joined #openstack-ansible | 18:16 | |
| *** hughmFLE_ has joined #openstack-ansible | 18:17 | |
| *** fops_ has quit IRC | 18:17 | |
| *** hughmFLEXin has quit IRC | 18:18 | |
| *** techstepper has joined #openstack-ansible | 18:19 | |
| *** thorst_ has joined #openstack-ansible | 18:21 | |
| *** dxiri_ has joined #openstack-ansible | 18:21 | |
| *** dxiri has quit IRC | 18:22 | |
| techstepper | Hey there. This is related to the issue that @todd_m_dav reported earlier today -- I've noticed that in the inventory file for the cinder_volumes containers, is_metal is set to false. But when the setup_hosts playbook runs, is_metal is set to true in the inventory. | 18:22 |
| *** fguillot has quit IRC | 18:22 | |
| techstepper | I do not know what's going on. | 18:22 |
| *** hughmFLEXin has joined #openstack-ansible | 18:23 | |
| *** chris_hultin is now known as chris_hultin|AWA | 18:24 | |
| *** gaudenz_ has quit IRC | 18:25 | |
| *** gaudenz has joined #openstack-ansible | 18:25 | |
| *** hughmFLE_ has quit IRC | 18:25 | |
| *** michauds has joined #openstack-ansible | 18:25 | |
| openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible-os_cinder: Add remote procedure call tunables to cinder https://review.openstack.org/360749 | 18:28 |
| cloudnull | techstepper: interesting | 18:32 |
| cloudnull | so you're saying that in the openstack_ansible_inventory.json file the is_metal flag is being set to False? | 18:33 |
| cloudnull | for cinder_volumens | 18:33 |
| cloudnull | and then being flipped when the os-cinder-install.yml play runs? | 18:33 |
| cloudnull | or is it simply being ignored? | 18:34 |
| cloudnull | also sorry, i may be missing some context. | 18:34 |
| *** spotz is now known as spotz_zzz | 18:35 | |
| *** cmart has joined #openstack-ansible | 18:40 | |
| techstepper | yes. I set it to false. And it's being flipped. | 18:41 |
| *** severion has joined #openstack-ansible | 18:42 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Update the neutron MTU settings on upgrade https://review.openstack.org/384635 | 18:42 |
| techstepper | Should I not consider an inventory file to be a source of truth? | 18:42 |
| *** afred312_ has joined #openstack-ansible | 18:42 | |
| *** derekjhyang_ has joined #openstack-ansible | 18:43 | |
| *** rackertom_ has joined #openstack-ansible | 18:43 | |
| *** ikalnytskyi_ has joined #openstack-ansible | 18:44 | |
| cloudnull | yes, the inventory file should be the source of truth | 18:44 |
| cloudnull | are you seeing it change in the inventory file or is it simply bieng ignored ? | 18:44 |
| *** raddaoui_ has joined #openstack-ansible | 18:45 | |
| cloudnull | also has the env.d file for cinder been changed to set cinder-volume is_metal: False? | 18:45 |
| techstepper | I am seeing it change in the inventory file. (I make the change to "false", run the playbook, look at the inventory again, it's set to true.) | 18:45 |
| *** ggillies_ has joined #openstack-ansible | 18:46 | |
| cloudnull | as noted here. http://docs.openstack.org/developer/openstack-ansible/developer-docs/extending.html | 18:46 |
| cloudnull | sorry, thats in relation to the env.d files | 18:46 |
| cloudnull | maybe palendae can expand on that but maybe its reindexing and pulling the value from env.d ? | 18:47 |
| techstepper | oh, it is set to true in /etc/openstack_deploy/env.d/cinder.yml. But why would that override what I explicitly put in the inventory? | 18:47 |
| cloudnull | im not 100% sure. env.d is used as the skeleton for all things related to inventory. my only guess is that it's reindexing that value. | 18:48 |
| techstepper | If the inventory is the source of truth, then explicit changes in that should supersede other things, or at least complain. | 18:48 |
| *** FrankZha- has joined #openstack-ansible | 18:49 | |
| *** dolphm_ has joined #openstack-ansible | 18:49 | |
| *** neillc has quit IRC | 18:49 | |
| *** odyssey4me_ has joined #openstack-ansible | 18:49 | |
| *** Aju has joined #openstack-ansible | 18:49 | |
| cloudnull | I know you can drop a new file in /etc/openstack_deploy/env.d to override what we have in the included inventory, you may want to give that a spin to see if it "fixes" the issue. | 18:49 |
| cloudnull | but I agree a defined value in the json file should be left alone or throw some sort of exception | 18:50 |
| *** irtermit- has joined #openstack-ansible | 18:50 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Automatically source the rc file for all Ansible commands https://review.openstack.org/386680 | 18:50 |
| *** _d34dh0r53_ has joined #openstack-ansible | 18:51 | |
| odyssey4me_ | cloudnull IIRC it was liberty or something | 18:51 |
| *** weezS has quit IRC | 18:51 | |
| cloudnull | tbt i've not worked on the inventory bits in some time and there's been a lot of changes there for this cycle so my general knowledge how it all works is a little lacking | 18:51 |
| cloudnull | odyssey4me_: say again ? | 18:51 |
| *** v1k0d3n has quit IRC | 18:51 | |
| *** jamesdenton has quit IRC | 18:51 | |
| *** Jack_Iv has quit IRC | 18:51 | |
| *** derekjhyang has quit IRC | 18:51 | |
| *** afred312 has quit IRC | 18:51 | |
| *** rackertom has quit IRC | 18:51 | |
| *** d34dh0r53 has quit IRC | 18:51 | |
| *** raddaoui has quit IRC | 18:51 | |
| *** irtermite has quit IRC | 18:51 | |
| *** odyssey4me has quit IRC | 18:51 | |
| *** palendae has quit IRC | 18:51 | |
| *** ikalnytskyi has quit IRC | 18:51 | |
| *** ggillies has quit IRC | 18:51 | |
| *** qwang has quit IRC | 18:51 | |
| *** afranc has quit IRC | 18:51 | |
| *** mgariepy has quit IRC | 18:51 | |
| *** FrankZhang has quit IRC | 18:51 | |
| *** _hanhart has quit IRC | 18:51 | |
| *** lbragstad has quit IRC | 18:51 | |
| *** dolphm has quit IRC | 18:51 | |
| *** mrhillsman has quit IRC | 18:51 | |
| *** dolphm_ is now known as dolphm | 18:51 | |
| *** odyssey4me_ is now known as odyssey4me | 18:51 | |
| cloudnull | oh there goes the neighborhood | 18:52 |
| *** derekjhyang_ is now known as derekjhyang | 18:52 | |
| odyssey4me | cloudnull http://eavesdrop.openstack.org/irclogs/%23openstack-ansible/%23openstack-ansible.2016-10-14.log.html#t2016-10-14T14:22:23 | 18:53 |
| *** rackertom_ is now known as rackertom | 18:53 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-tests: Fix package name for python-pyasn1 on CentOS https://review.openstack.org/386740 | 18:53 |
| *** ikalnytskyi_ is now known as ikalnytskyi | 18:53 | |
| cloudnull | ah. | 18:53 |
| *** raddaoui_ is now known as raddaoui | 18:53 | |
| cloudnull | techstepper: is this a liberty deployment? | 18:53 |
| *** lbragstad has joined #openstack-ansible | 18:53 | |
| *** fops has quit IRC | 18:54 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Automatically source the rc file for all Ansible commands https://review.openstack.org/386680 | 18:55 |
| *** fops has joined #openstack-ansible | 18:55 | |
| odyssey4me | cloudnull ^ hopefully that's just a bit less horrible and a bit more nice :) | 18:55 |
| *** adrian_otto has quit IRC | 18:56 | |
| *** Jack_Iv has joined #openstack-ansible | 18:56 | |
| *** thorst_ has quit IRC | 18:57 | |
| *** neillc has joined #openstack-ansible | 18:57 | |
| *** thorst_ has joined #openstack-ansible | 18:58 | |
| *** Mudpuppy has quit IRC | 19:00 | |
| *** mrhillsman has joined #openstack-ansible | 19:00 | |
| *** _hanhart_ has joined #openstack-ansible | 19:00 | |
| odyssey4me | ok, I'm out agian | 19:00 |
| *** qwang has joined #openstack-ansible | 19:03 | |
| *** fops has quit IRC | 19:03 | |
| cloudnull | odyssey4me: i dont think we want to include the vars on all ansible commands. | 19:03 |
| cloudnull | maybe just ansible || ansible-playbooks | 19:03 |
| cloudnull | in the case of ansible-vault we'd be including the secretes file which is encrypted and it breaks the command. | 19:04 |
| *** johnmilton has quit IRC | 19:06 | |
| *** thorst_ has quit IRC | 19:06 | |
| *** _hanhart_ is now known as _hanhart | 19:09 | |
| *** mcarden has quit IRC | 19:09 | |
| *** erikmwilson has quit IRC | 19:09 | |
| *** neith has quit IRC | 19:10 | |
| *** mcarden has joined #openstack-ansible | 19:11 | |
| *** karimb has quit IRC | 19:12 | |
| *** neith has joined #openstack-ansible | 19:12 | |
| *** severion has quit IRC | 19:13 | |
| *** v1k0d3n has joined #openstack-ansible | 19:13 | |
| *** hughmFLEXin has quit IRC | 19:14 | |
| techstepper | hmm. can't work on it right now -- our test lab has been powered down. | 19:15 |
| techstepper | @cloudnull: It's a mitaka deployment. | 19:15 |
| *** ableto has joined #openstack-ansible | 19:16 | |
| *** erikmwilson has joined #openstack-ansible | 19:16 | |
| cloudnull | ok. | 19:17 |
| *** Guest37578 has joined #openstack-ansible | 19:17 | |
| cloudnull | i dont have a mitaka test env up right now so i cant test but i'd be curious what the root cause is there and if we see the same behviour in newton. | 19:17 |
| *** kylek3h is now known as kylek3h_away | 19:17 | |
| *** hybridpollo has joined #openstack-ansible | 19:21 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server: Move systemd handler above restart handler https://review.openstack.org/386604 | 19:25 |
| *** meteorfox has quit IRC | 19:29 | |
| *** mubix_ has quit IRC | 19:29 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift: Fix swift init scripts w/o dedicated replication https://review.openstack.org/386679 | 19:30 |
| *** mubix_ has joined #openstack-ansible | 19:30 | |
| *** meteorfox has joined #openstack-ansible | 19:31 | |
| *** mgariepy_ has joined #openstack-ansible | 19:34 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_horizon: Add conditional for multi-domain support https://review.openstack.org/386753 | 19:35 |
| *** cmart has quit IRC | 19:36 | |
| *** mgariepy_ has quit IRC | 19:37 | |
| *** mgariepy has joined #openstack-ansible | 19:37 | |
| openstackgerrit | Steve Lewis (stevelle) proposed openstack/openstack-ansible-os_ceilometer: Update ceilometer app.wsgi template per upstream https://review.openstack.org/385697 | 19:38 |
| *** thorst_ has joined #openstack-ansible | 19:38 | |
| *** spotz_zzz is now known as spotz | 19:41 | |
| *** thorst_ has quit IRC | 19:45 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create: Drop container setup script independent of container_networks https://review.openstack.org/386699 | 19:46 |
| *** thorst_ has joined #openstack-ansible | 19:46 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create: Don't fail veth-cleanup template when no container_networks https://review.openstack.org/386700 | 19:46 |
| *** jamesdenton has joined #openstack-ansible | 19:50 | |
| *** cmart has joined #openstack-ansible | 19:51 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Allow isolated venv creation https://review.openstack.org/386759 | 19:54 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic: Configure ironic to use keystone v2 https://review.openstack.org/386234 | 19:54 |
| *** thorst_ has quit IRC | 19:54 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_barbican: Remove duplicate and unnecessary pip packages https://review.openstack.org/386318 | 19:58 |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican: Use barbican-manage https://review.openstack.org/386312 | 20:00 |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican: Update paste, policy and rootwrap configurations 2016-10-13 https://review.openstack.org/386276 | 20:01 |
| *** haad1 has joined #openstack-ansible | 20:01 | |
| *** adrian_otto has joined #openstack-ansible | 20:01 | |
| *** todd_m_dav has quit IRC | 20:02 | |
| *** cjmarti has joined #openstack-ansible | 20:02 | |
| *** castulo has quit IRC | 20:03 | |
| *** cjmarti is now known as castulo | 20:03 | |
| *** luzC has quit IRC | 20:03 | |
| *** luzC has joined #openstack-ansible | 20:03 | |
| *** Aju has quit IRC | 20:05 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests: Fix neutron_agent IP to be neutron_server IP https://review.openstack.org/386530 | 20:07 |
| *** rromans has quit IRC | 20:08 | |
| *** techstepper has quit IRC | 20:11 | |
| *** rrrobbb has joined #openstack-ansible | 20:12 | |
| *** ableto has quit IRC | 20:12 | |
| *** castulo has quit IRC | 20:12 | |
| *** Mudpuppy has joined #openstack-ansible | 20:13 | |
| *** castulo has joined #openstack-ansible | 20:13 | |
| odyssey4me | cloudnull ok, I think we'll want it for ansible-galaxy too, so the roles end up in the right place - lemme fix that up | 20:13 |
| *** Mudpuppy_ has joined #openstack-ansible | 20:13 | |
| *** javeriak has quit IRC | 20:17 | |
| *** Mudpuppy has quit IRC | 20:17 | |
| *** pcaruana has quit IRC | 20:18 | |
| agrebennikov | hey odyssey4me very quick question - how do we submit specs for the roles? | 20:20 |
| agrebennikov | which repo | 20:20 |
| agrebennikov | so that we can get it here http://specs.openstack.org/openstack/openstack-ansible-specs/ | 20:20 |
| *** jamesdenton has quit IRC | 20:22 | |
| *** castulo has quit IRC | 20:23 | |
| *** castulo has joined #openstack-ansible | 20:24 | |
| *** luzC- has joined #openstack-ansible | 20:24 | |
| *** luzC has quit IRC | 20:24 | |
| *** luzC- is now known as luzC | 20:25 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift: Use dictionary for service mappings https://review.openstack.org/386635 | 20:25 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-repo_build: Ensure the venv creation process is able to work with isolated deps https://review.openstack.org/386765 | 20:25 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create: Don't fail veth-cleanup template when no container_networks https://review.openstack.org/386766 | 20:26 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create: Drop container setup script independent of container_networks https://review.openstack.org/386767 | 20:26 |
| cloudnull | agrebennikov: you'll want to pull down the specs repo and commit it there. | 20:26 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server: Move systemd handler above restart handler https://review.openstack.org/386768 | 20:26 |
| cloudnull | https://github.com/openstack/openstack-ansible-specs | 20:27 |
| cloudnull | agrebennikov: ^ | 20:27 |
| agrebennikov | aaahhhh! | 20:27 |
| agrebennikov | pretty obvious)) | 20:27 |
| cloudnull | no worries. | 20:27 |
| cloudnull | there are a lot of repos to sift through | 20:27 |
| cloudnull | :) | 20:27 |
| *** rmelero has joined #openstack-ansible | 20:27 | |
| agrebennikov | cloudnull, because I was looking at https://github.com/openstack/openstack-ansible-os_monasca/tree/master/doc/source | 20:28 |
| *** rmelero has quit IRC | 20:28 | |
| agrebennikov | rmelero, https://github.com/openstack/openstack-ansible-specs this is what we need)) | 20:28 |
| odyssey4me | agrebennikov yes, that's the specs repo | 20:28 |
| *** rmelero has joined #openstack-ansible | 20:28 | |
| agrebennikov | odyssey4me, cloudnull so what should be commited to those <repo>/docs/source? | 20:29 |
| cloudnull | agrebennikov: ah yes. we aggregate the docs and release notes but the specs in a self contained repo. | 20:29 |
| agrebennikov | I see other projects like keystone having rsts in those folders | 20:29 |
| odyssey4me | agrebennikov specs are in https://github.com/openstack/openstack-ansible-specs/tree/master/specs | 20:29 |
| odyssey4me | there's a template there | 20:29 |
| cloudnull | agrebennikov: documentation changes that are specific to that role will be in that dir | 20:30 |
| odyssey4me | oh, you mean generally in all the repositories? | 20:30 |
| agrebennikov | yeah | 20:30 |
| odyssey4me | ^ what cloudnull said | 20:30 |
| agrebennikov | jfyi http://specs.openstack.org/openstack/openstack-ansible-specs/specs/template.rst points to nowhere | 20:30 |
| agrebennikov | probably this is why we were confused | 20:30 |
| cloudnull | Matias: you around? | 20:31 |
| cloudnull | looking at https://bugs.launchpad.net/openstack-ansible/+bug/1624791 | 20:31 |
| openstack | Launchpad bug 1624791 in openstack-ansible "Horizon randomly fails to connect to the service APIs" [High,New] | 20:31 |
| cloudnull | curious if you still see this with post RC releases? | 20:31 |
| cloudnull | ** post RC1 releases ** | 20:31 |
| odyssey4me | agrebennikov it should point to https://github.com/openstack/openstack-ansible-specs/blob/master/specs/template.rst | 20:32 |
| odyssey4me | a patch to fix that would be appreciated | 20:32 |
| agrebennikov | :) | 20:32 |
| agrebennikov | if I new what handles the page... | 20:32 |
| odyssey4me | agrebennikov https://raw.githubusercontent.com/openstack/openstack-ansible-specs/master/doc/source/index.rst | 20:33 |
| odyssey4me | looks like a relative path issue | 20:33 |
| agrebennikov | yeah | 20:33 |
| *** retreved has quit IRC | 20:33 | |
| odyssey4me | ok, I'm out again - have a great w/end all | 20:33 |
| agrebennikov | have a great time)) | 20:33 |
| agrebennikov | I'll fix it | 20:33 |
| *** retreved has joined #openstack-ansible | 20:34 | |
| *** gouthamr_ is now known as gouthamr | 20:35 | |
| *** retreved has quit IRC | 20:38 | |
| cloudnull | later odyssey4me | 20:43 |
| *** thorst_ has joined #openstack-ansible | 20:44 | |
| *** cmart has quit IRC | 20:44 | |
| *** asettle has joined #openstack-ansible | 20:44 | |
| cloudnull | ha. odyssey4me still around? | 20:44 |
| openstackgerrit | Andrey Grebennikov proposed openstack/openstack-ansible-specs: Fixed the link to the spec template https://review.openstack.org/386772 | 20:45 |
| cloudnull | on https://bugs.launchpad.net/openstack-ansible/+bug/1633438 | 20:45 |
| openstack | Launchpad bug 1633438 in openstack-ansible "Handle intentional task failures properly" [Undecided,New] | 20:45 |
| cloudnull | -- are we looking to swap out ignore_errors: true for failed_when: false | 20:45 |
| *** jamesdenton has joined #openstack-ansible | 20:45 | |
| *** thorst_ has quit IRC | 20:48 | |
| cloudnull | agrebennikov: i think that can be a relative path if uses "../../specs/template.rst" | 20:48 |
| cloudnull | it would seem the symlink is not working though | 20:49 |
| cloudnull | if you're needing to change it that is | 20:49 |
| agrebennikov | cloudnull, seems neither of the paths work :/ I tried all other combinations of http://specs.openstack.org/openstack/openstack-ansible-specs/...... | 20:50 |
| agrebennikov | I mean I have no idea how the nginx (or whetever the web server is is) is handling links | 20:50 |
| cloudnull | ah. you know this is likely not working outside of the doc root | 20:51 |
| agrebennikov | well..... HARDCODE everything :D | 20:51 |
| agrebennikov | then the template probably has to be moved somewhere else? | 20:51 |
| cloudnull | hahahaa. | 20:52 |
| cloudnull | idk tbh. | 20:52 |
| agrebennikov | I mean some upper folders | 20:52 |
| cloudnull | looking at the rest of the index it seems to be working... | 20:53 |
| cloudnull | which may be functional because of the .. toctree:: call | 20:53 |
| agrebennikov | oh.. I have no idea what is is :/ | 20:54 |
| cloudnull | maybe we need to do something like so | 20:54 |
| cloudnull | http://cdn.pasteraw.com/kpzt1so6452rbjtcaunp0d5d3lc6la4 | 20:54 |
| cloudnull | and move template.rst into a dir | 20:55 |
| cloudnull | which we know will work per the rest of specs | 20:55 |
| agrebennikov | yeah.... probably | 20:56 |
| cloudnull | yea it looks like the glob call with that doc magic function pulls in the files into the specs dir | 20:56 |
| cloudnull | which it doesn't otherwise. | 20:56 |
| cloudnull | so http://specs.openstack.org/openstack/openstack-ansible-specs/specs/template.rst is 404 | 20:56 |
| cloudnull | nonsense! i tell ya. | 20:57 |
| cloudnull | HARDCODE everything! | 20:57 |
| cloudnull | :) | 20:57 |
| agrebennikov | exactly))) this is why raised the question "how the hell we submit the rst")) | 20:57 |
| cloudnull | I'd be for moving it into a dir and should we ever want to make more templates for different things we can. | 20:58 |
| *** galstrom is now known as galstrom_zzz | 20:58 | |
| *** cmart has joined #openstack-ansible | 20:58 | |
| agrebennikov | ok, I'll do it | 20:58 |
| agrebennikov | cloudnull, hm... but in this case it's not going to point to github page :/ | 21:01 |
| *** hughmFLEXin has joined #openstack-ansible | 21:02 | |
| cloudnull | right the template should added to http://specs.openstack.org/openstack/openstack-ansible-specs/specs/templates/template.html | 21:03 |
| *** jamesdenton has quit IRC | 21:04 | |
| cloudnull | oh... | 21:04 |
| cloudnull | so http://specs.openstack.org/openstack/openstack-ansible-specs/specs/template.html | 21:04 |
| cloudnull | i guess our issue is that its the extension | 21:05 |
| agrebennikov | what odyssey4me said - it should point to https://github.com/openstack/openstack-ansible-specs/blob/master/specs/template.rst | 21:05 |
| openstackgerrit | Andrey Grebennikov proposed openstack/openstack-ansible-specs: Fixed the link to the spec template https://review.openstack.org/386772 | 21:06 |
| agrebennikov | maybe something like this? | 21:06 |
| cloudnull | so then maybe we point it at http://git.openstack.org/cgit/openstack/openstack-ansible-specs/tree/specs/template.rst | 21:06 |
| *** weezS has joined #openstack-ansible | 21:06 | |
| *** cjmarti has joined #openstack-ansible | 21:06 | |
| *** michauds has quit IRC | 21:06 | |
| cloudnull | if we want it to point to the raw rst file | 21:06 |
| openstackgerrit | Kyle L. Henderson proposed openstack/openstack-ansible-galera_server: Add tests/common to .gitignore https://review.openstack.org/386776 | 21:07 |
| cloudnull | agrebennikov: yup that should work | 21:07 |
| *** castulo has quit IRC | 21:07 | |
| *** cjmarti is now known as castulo | 21:07 | |
| agrebennikov | but we need to try and see how it will look like first | 21:07 |
| agrebennikov | maybe it's ugly)) | 21:07 |
| agrebennikov | or | 21:08 |
| agrebennikov | maybe it'll not work also - other specs are html :/ | 21:08 |
| cloudnull | yea. the html makes sense to m e | 21:08 |
| cloudnull | *me | 21:08 |
| agrebennikov | aahhh, so that toctree does magic)) | 21:09 |
| agrebennikov | it turns rst to html somehow | 21:09 |
| cloudnull | yea it looks like it | 21:09 |
| agrebennikov | ok | 21:09 |
| agrebennikov | I'd also extend the section with the link to the raw template if I new how to operate with this language ;/ | 21:10 |
| agrebennikov | *knew | 21:10 |
| agrebennikov | anyway, let's see. maybe others have another opinion | 21:10 |
| agrebennikov | tnx for your help)) | 21:10 |
| cloudnull | anytime | 21:11 |
| cloudnull | i too have no idea how doc things work. | 21:11 |
| cloudnull | asettle: ^ | 21:11 |
| cloudnull | asettle: is the resident doc expert | 21:11 |
| *** javeriak has joined #openstack-ansible | 21:17 | |
| *** javeriak has quit IRC | 21:22 | |
| *** jheroux has quit IRC | 21:23 | |
| *** adrian_otto has quit IRC | 21:23 | |
| *** adrian_otto has joined #openstack-ansible | 21:25 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum: Update paste, policy and rootwrap configurations 2016-10-12 https://review.openstack.org/385546 | 21:27 |
| *** haad1 has quit IRC | 21:29 | |
| *** adrian_otto has quit IRC | 21:30 | |
| *** adrian_otto has joined #openstack-ansible | 21:32 | |
| *** adrian_otto has quit IRC | 21:35 | |
| * asettle peaks in | 21:37 | |
| asettle | cloudnull: yes? | 21:37 |
| *** Jack_Iv has quit IRC | 21:39 | |
| *** adrian_otto has joined #openstack-ansible | 21:40 | |
| *** adrian_otto has quit IRC | 21:42 | |
| cloudnull | was just refering to doc generation magic earlier | 21:43 |
| cloudnull | specifically https://review.openstack.org/386772 | 21:43 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386792 | 21:43 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ceph_client: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386793 | 21:43 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_client: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386794 | 21:43 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-galera_server: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386795 | 21:44 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-haproxy_server: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386796 | 21:44 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386797 | 21:44 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386798 | 21:44 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-openstack_hosts: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386799 | 21:44 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_aodh: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386800 | 21:44 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_barbican: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386801 | 21:44 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_ceilometer: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386802 | 21:44 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_cinder: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386803 | 21:44 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_nova: Remove nova-cert https://review.openstack.org/386804 | 21:44 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_cloudkitty: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386805 | 21:45 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_designate: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386806 | 21:45 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_glance: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386807 | 21:45 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_gnocchi: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386808 | 21:45 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_heat: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386809 | 21:45 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_horizon: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386810 | 21:45 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_ironic: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386811 | 21:45 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_keystone: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386812 | 21:45 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_magnum: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386813 | 21:45 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_neutron: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386814 | 21:45 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_nova: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386815 | 21:46 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_rally: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386816 | 21:46 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_sahara: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386818 | 21:46 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_swift: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386819 | 21:46 |
| *** rrrobbb has quit IRC | 21:46 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_tempest: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386821 | 21:46 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_trove: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386822 | 21:46 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_watcher: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386823 | 21:46 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_zaqar: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386824 | 21:46 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-pip_install: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386825 | 21:46 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-pip_lock_down: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386826 | 21:46 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-rabbitmq_server: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386827 | 21:47 |
| *** asettle has quit IRC | 21:49 | |
| *** rrrobbb has joined #openstack-ansible | 21:50 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible: Remove nova-cert environment information https://review.openstack.org/386829 | 21:50 |
| *** rrrobbb has quit IRC | 21:51 | |
| *** retreved has joined #openstack-ansible | 21:52 | |
| *** retreved has quit IRC | 21:53 | |
| *** thorst_ has joined #openstack-ansible | 21:53 | |
| cloudnull | nice jmccrory! | 21:54 |
| cloudnull | it'll be good to see that thing gone. | 21:54 |
| jmccrory | cloudnull: yeah, it's about time | 21:54 |
| jmccrory | one less useless container | 21:54 |
| cloudnull | indeed . | 21:54 |
| rmelero | Submitted a new spec. https://github.com/openstack/openstack-ansible-specs/pull/1/commits/0d75e6acde2081ad1aab7db02de463fd2ab980c4 | 21:55 |
| rmelero | anything else for me to do? | 21:55 |
| cloudnull | I'd also like to see nova_api_os_compute nova_api_metadata become one . | 21:55 |
| cloudnull | idk if others want that but it'd like it :) | 21:55 |
| cloudnull | rmelero: we need that to be submitted through git-review | 21:56 |
| cloudnull | the PRs against github are autoclosed and generally ignored . | 21:56 |
| rmelero | ah ok so scrap that pull request. | 21:56 |
| cloudnull | sorry | 21:56 |
| rmelero | np | 21:56 |
| rmelero | my hamster ate my homework instructions | 21:57 |
| cloudnull | it happens to me all the time :) | 21:57 |
| *** javeriak has joined #openstack-ansible | 21:58 | |
| *** jmckind has quit IRC | 21:59 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-repo_build: Ensure the venv creation process is able to work with isolated deps https://review.openstack.org/386765 | 22:01 |
| *** thorst_ has quit IRC | 22:02 | |
| *** jperry has quit IRC | 22:06 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-lxc_container_create: Use ping module instead of wait_for to check container https://review.openstack.org/386835 | 22:09 |
| *** sdake_ has quit IRC | 22:14 | |
| *** rromans has joined #openstack-ansible | 22:15 | |
| *** michaelgugino has quit IRC | 22:19 | |
| openstackgerrit | Ramon Melero proposed openstack/openstack-ansible-specs: monasca-ha & monasca-agent roles https://review.openstack.org/386839 | 22:20 |
| cmart | is anyone else regularly seeing Ansible hang for a long time (20+ minutes) during group creation tasks? on the target host there will be a `/usr/sbin/groupadd` process running. there will also be an /etc/group.lock and an /etc/gshadow.lock referring to that process. I'm consistently seeing it during creation of the haproxy group, cinder group, etc. | 22:22 |
| logan- | which branch cmart | 22:23 |
| cmart | newton/stable. pulled on Wednesday | 22:23 |
| logan- | no i'm not seeing anything like that on newton | 22:23 |
| rmelero | strace the groupadd process. could provide clues to where it's hanging. | 22:23 |
| cmart | will do | 22:24 |
| logan- | there's a big wait during the swift playbook on xenial that is going to be resolved soon, but it is during service restart tasks.. so beware of that for now | 22:24 |
| cmart | i should mention that it does eventually complete, after 30 minutes or so | 22:25 |
| *** cathrichardson has quit IRC | 22:27 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible: Remove nova-cert environment information https://review.openstack.org/386829 | 22:33 |
| cmart | logan- strace shows that my groupadd process keeps trying (and failing) to talk to my LDAP server. so it's likely an issue unique to me. nothing to see here, move along :) | 22:36 |
| cmart | https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=498635 that's the issue. nothing OSA related | 22:37 |
| openstack | Debian bug 498635 in libnss-ldap "passwd: useradd -r and groupadd -r don't play nicely with ldap" [Important,Fixed] | 22:37 |
| *** sdake has joined #openstack-ansible | 22:44 | |
| cloudnull | logan-: inorder to make https://review.openstack.org/#/c/386835/ go i think we'll need to tune up the inventory to trigger container handling. | 22:47 |
| cloudnull | the minimal inventory in tests simply treats everything like a standard host. | 22:47 |
| cloudnull | btw did you ever get the local plugins test to work ? | 22:48 |
| *** hughmFLEXin has quit IRC | 22:49 | |
| *** javeriak has quit IRC | 22:50 | |
| *** hughmFLEXin has joined #openstack-ansible | 22:51 | |
| *** hughmFLEXin has joined #openstack-ansible | 22:51 | |
| *** gouthamr has quit IRC | 22:56 | |
| *** sdake has quit IRC | 22:57 | |
| *** sdake has joined #openstack-ansible | 22:58 | |
| *** sdake has quit IRC | 22:58 | |
| *** thorst has joined #openstack-ansible | 22:59 | |
| *** askb has joined #openstack-ansible | 23:01 | |
| *** askb has quit IRC | 23:01 | |
| *** Mudpuppy_ has quit IRC | 23:03 | |
| *** sdake has joined #openstack-ansible | 23:06 | |
| *** thorst has quit IRC | 23:07 | |
| *** mgagne_ is now known as mgagne | 23:12 | |
| *** jcrst has quit IRC | 23:13 | |
| dxiri_ | hello everyone! I am following this page to do a test deployment: http://docs.openstack.org/developer/openstack-ansible/install-guide/app-config-test.html | 23:13 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Add an option to generate fixed MAC addresses https://review.openstack.org/385800 | 23:14 |
| dxiri_ | and have a question about the "Host Network configuration" | 23:14 |
| dxiri_ | where should I put that? | 23:14 |
| cloudnull | evrardjp logan- https://review.openstack.org/385800 -- if you have a moment. that should address all of the concerns and make the fixed mac address capability optional. | 23:14 |
| cloudnull | jmccrory: ^ would love you opinion on that too. | 23:15 |
| dxiri_ | I already edited the deployment configuration for the environment layout, but no idea where that host configuration goes | 23:15 |
| cloudnull | dxiri_: the host networking is not something that OSA does for you. | 23:15 |
| cloudnull | the file in the example is /etc/network/interfaces | 23:15 |
| dxiri_ | ahhhhhh | 23:15 |
| dxiri_ | now it makes sense :) | 23:16 |
| dxiri_ | thank you | 23:16 |
| dxiri_ | doesn't say that anywhere on the doc so it would be good to put it in | 23:16 |
| cloudnull | in general with my deployments I put all of my osa specific networking bits into a file under /etc/network/interfaces.d | 23:16 |
| cloudnull | ++ that would be good. | 23:16 |
| cloudnull | if this is an AIO | 23:16 |
| cloudnull | we do generate a config | 23:17 |
| cloudnull | but its AIO specific | 23:17 |
| dxiri_ | it isn't | 23:17 |
| cloudnull | ok. | 23:17 |
| dxiri_ | already played with AIO | 23:17 |
| dxiri_ | now jumping into 3 hosts (appendix A) example | 23:17 |
| cloudnull | nice | 23:17 |
| dxiri_ | just needed a few tweaks for the network I have on the lab | 23:17 |
| dxiri_ | is a /24 enough for the mgmt network? | 23:18 |
| dxiri_ | cause that's what we have available here | 23:18 |
| cloudnull | that should be enough. | 23:18 |
| rmelero | more than enough. don't forget to exclude the ips of the physical hosts tho. | 23:18 |
| cloudnull | our inventory process should already exclude host addresses. | 23:19 |
| cloudnull | rmelero: if it isn't that's bug we should fix. | 23:19 |
| rmelero | hmm. don't have access to the environment anymore, but i had to manually add some ips to prevent ip conflicts. prob a misconfig on my part. | 23:20 |
| dxiri_ | yep I did, I have a few hosts with IPs on the middle of the range I needed to exclude | 23:20 |
| dxiri_ | that + the gateway, and the first 15 addresses | 23:20 |
| dxiri_ | the range I specified it like this: - "172.25.190.1,172.25.190.14" | 23:21 |
| dxiri_ | hopefully that is ok | 23:21 |
| *** sdake has quit IRC | 23:21 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Enable fixed mac address generation https://review.openstack.org/386847 | 23:23 |
| cloudnull | that looks good. | 23:24 |
| logan- | looking thru it cloudnull | 23:24 |
| logan- | could you drop lxc_container_fixed_mac in defaults too just so its visible without digging thru tasks | 23:25 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Add an option to generate fixed MAC addresses https://review.openstack.org/385800 | 23:26 |
| cloudnull | yup. i had forgotten that. | 23:26 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Add an option to generate fixed MAC addresses https://review.openstack.org/385800 | 23:27 |
| cloudnull | sorry for the late edit | 23:27 |
| cloudnull | i forgot the conditional in one other place. | 23:27 |
| *** afranc has joined #openstack-ansible | 23:28 | |
| logan- | didn't know you could use set_facts that way | 23:32 |
| *** hughmFLEXin has quit IRC | 23:33 | |
| cloudnull | its useful when you need to set something dynamically | 23:33 |
| logan- | yep. ive wanted to do that a few times before. good to know :) | 23:33 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_container_create: Add an option to generate fixed MAC addresses https://review.openstack.org/385800 | 23:38 |
| cloudnull | and now there's a test | 23:38 |
| cloudnull | ok . i've got to run | 23:38 |
| cloudnull | bbl | 23:38 |
| cloudnull | thanks for giving that a look and all your feedback on it so far. | 23:38 |
| logan- | no problem! ttyl | 23:40 |
| *** weezS has quit IRC | 23:47 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
