| *** dfflanders has joined #openstack-ansible | 00:02 | |
| *** williamcaban has quit IRC | 00:04 | |
| *** williamcaban has joined #openstack-ansible | 00:05 | |
| *** rmelero has quit IRC | 00:07 | |
| *** ikalnytskyi_ has joined #openstack-ansible | 00:08 | |
| *** mubix has quit IRC | 00:08 | |
| *** chris_hultin|AWA has quit IRC | 00:08 | |
| *** mubix_ has joined #openstack-ansible | 00:09 | |
| *** ikalnytskyi has quit IRC | 00:09 | |
| *** ikalnytskyi_ is now known as ikalnytskyi | 00:09 | |
| *** aludwar has quit IRC | 00:09 | |
| *** chris_hultin|AWA has joined #openstack-ansible | 00:10 | |
| *** batkins61 has quit IRC | 00:10 | |
| *** chris_hultin|AWA is now known as chris_hultin | 00:10 | |
| *** aludwar has joined #openstack-ansible | 00:11 | |
| *** gouthamr has joined #openstack-ansible | 00:16 | |
| *** chris_hultin has quit IRC | 00:19 | |
| *** adrian_otto has joined #openstack-ansible | 00:21 | |
| *** sdake_ is now known as sdake | 00:22 | |
| *** gouthamr has quit IRC | 00:23 | |
| *** gouthamr has joined #openstack-ansible | 00:23 | |
| *** aetaric_ has quit IRC | 00:23 | |
| *** toan has quit IRC | 00:24 | |
| *** b3rn-n00dl3s has quit IRC | 00:24 | |
| *** b3rn-n00dl3s has joined #openstack-ansible | 00:25 | |
| *** chris_hultin|AWA has joined #openstack-ansible | 00:28 | |
| *** chris_hultin|AWA is now known as chris_hultin | 00:28 | |
| *** markvoelker has quit IRC | 00:28 | |
| *** toan has joined #openstack-ansible | 00:29 | |
| *** aetaric has joined #openstack-ansible | 00:29 | |
| *** adrian_otto has quit IRC | 00:29 | |
| *** adrian_otto has joined #openstack-ansible | 00:31 | |
| *** adrian_otto has quit IRC | 00:33 | |
| *** adrian_otto has joined #openstack-ansible | 00:36 | |
| *** markvoelker has joined #openstack-ansible | 00:36 | |
| *** mubix_ has quit IRC | 00:38 | |
| *** mubix_ has joined #openstack-ansible | 00:40 | |
| *** mgagne has quit IRC | 00:40 | |
| *** cmart has quit IRC | 00:42 | |
| *** adrian_otto has quit IRC | 00:42 | |
| *** mgagne has joined #openstack-ansible | 00:45 | |
| *** sdake has quit IRC | 00:46 | |
| *** jamielennox is now known as jamielennox|away | 00:56 | |
| *** aetaric has quit IRC | 00:59 | |
| *** sdake has joined #openstack-ansible | 01:01 | |
| *** jamielennox|away is now known as jamielennox | 01:02 | |
| *** thorst has joined #openstack-ansible | 01:04 | |
| *** aetaric has joined #openstack-ansible | 01:05 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_nova: Use run_once to create authorized keys file script https://review.openstack.org/385221 | 01:05 |
|---|---|---|
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_nova: Use run_once to create authorized keys file script https://review.openstack.org/385221 | 01:06 |
| *** dfflanders has quit IRC | 01:07 | |
| *** chris_hultin has quit IRC | 01:09 | |
| *** chris_hultin|AWA has joined #openstack-ansible | 01:17 | |
| *** javeriak has joined #openstack-ansible | 01:17 | |
| *** chris_hultin|AWA is now known as chris_hultin | 01:17 | |
| openstackgerrit | Merged openstack/openstack-ansible: Fix value for openstack_host_manage_hosts_file https://review.openstack.org/385089 | 01:18 |
| *** jcrst has quit IRC | 01:20 | |
| *** thorst has quit IRC | 01:22 | |
| *** javeriak has quit IRC | 01:27 | |
| *** Mudpuppy has joined #openstack-ansible | 01:28 | |
| *** dfflanders has joined #openstack-ansible | 01:37 | |
| *** jcrst has joined #openstack-ansible | 01:38 | |
| openstackgerrit | Merged openstack/openstack-ansible: Move inventory generation code into lib https://review.openstack.org/383159 | 01:45 |
| *** woodard has joined #openstack-ansible | 01:47 | |
| *** woodard has quit IRC | 01:50 | |
| *** woodard has joined #openstack-ansible | 01:51 | |
| *** phalmos_ has joined #openstack-ansible | 01:55 | |
| *** phalmos has quit IRC | 01:55 | |
| *** thorst has joined #openstack-ansible | 01:58 | |
| *** thorst has quit IRC | 02:03 | |
| *** dfflanders has quit IRC | 02:09 | |
| *** jamesdenton has joined #openstack-ansible | 02:11 | |
| *** woodard has quit IRC | 02:14 | |
| *** rmelero has joined #openstack-ansible | 02:26 | |
| *** javeriak has joined #openstack-ansible | 02:28 | |
| *** rmelero has quit IRC | 02:36 | |
| *** rmelero has joined #openstack-ansible | 02:37 | |
| *** jcrst has quit IRC | 02:41 | |
| *** Mudpuppy has quit IRC | 02:41 | |
| *** kjw3 has quit IRC | 02:41 | |
| *** Mudpuppy has joined #openstack-ansible | 02:42 | |
| *** Mudpuppy has quit IRC | 02:46 | |
| *** dragonmaster has joined #openstack-ansible | 02:47 | |
| *** shan5464_ has quit IRC | 02:47 | |
| *** palendae has quit IRC | 02:49 | |
| *** palendae has joined #openstack-ansible | 02:50 | |
| *** palendae is now known as Guest34339 | 02:50 | |
| *** shananigans has joined #openstack-ansible | 02:52 | |
| *** Guest34339 is now known as palendae | 02:52 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Update the neutron MTU settings on upgrade https://review.openstack.org/384635 | 02:56 |
| *** sdake has quit IRC | 02:57 | |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install: Ensure that package installs can fall back https://review.openstack.org/384966 | 03:00 |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install: Ensure that package installs can fall back https://review.openstack.org/384969 | 03:00 |
| mattt | cloudnull: tried that again, not sure it worked | 03:08 |
| cloudnull | mattt: the script? | 03:08 |
| cloudnull | for mtu changes? | 03:08 |
| cloudnull | is it the script or it it not executing ? | 03:09 |
| mattt | cloudnull: i didn't catch a glimpse of the script before it got cleaned up, but my vxlan network didn't get an MTU of 1450 | 03:09 |
| *** jamesdenton has quit IRC | 03:09 | |
| cloudnull | what got set? | 03:09 |
| cloudnull | nothing? | 03:09 |
| mattt | just the public network | 03:10 |
| mattt | let me re-run it and disable script from getting deleted | 03:10 |
| mattt | brb | 03:10 |
| *** gouthamr has quit IRC | 03:10 | |
| mattt | oh, it doesn't get cleaned up | 03:11 |
| *** aetaric has quit IRC | 03:12 | |
| cloudnull | interesting. i see the same on a quite test. | 03:13 |
| cloudnull | poking at it now | 03:13 |
| *** palendae has quit IRC | 03:14 | |
| *** odyssey4me has quit IRC | 03:14 | |
| mattt | cloudnull: ok, so here's what i get | 03:14 |
| mattt | cloudnull: https://gist.github.com/mattt416/c1566302d19912846bb652714c330492 ... the vxlan network has physical_network=NULL in the db | 03:14 |
| *** odyssey4me has joined #openstack-ansible | 03:15 | |
| mattt | cloudnull: network_type=vxlan, physical_network=NULL | 03:15 |
| *** aetaric has joined #openstack-ansible | 03:15 | |
| *** palendae has joined #openstack-ansible | 03:16 | |
| *** palendae is now known as Guest35069 | 03:16 | |
| cloudnull | so for a physical_network=NULL we should just assume 1500 ? | 03:18 |
| cloudnull | or 1450 ? | 03:18 |
| *** sdake has joined #openstack-ansible | 03:18 | |
| mattt | cloudnull: refresh that gist, i dumped the mysql table in there so you can see | 03:19 |
| *** jamielennox is now known as jamielennox|away | 03:20 | |
| cloudnull | ok so if its NULL I think we should just assume its a tenant type network and use 1450 | 03:21 |
| *** Guest35069 has quit IRC | 03:21 | |
| cloudnull | I have an AIO up that looks like so http://cdn.pasteraw.com/jjyea2e5ow6b3n9cugl6reg5w86pakp | 03:21 |
| cloudnull | an both the ha net and PRIVAATE are based on tenant nets | 03:22 |
| cloudnull | so that makes sense. | 03:22 |
| *** palendae_ has joined #openstack-ansible | 03:22 | |
| *** palendae_ is now known as Guest55355 | 03:23 | |
| mattt | cloudnull: yeah, i can't really suggest which way to go here, i'm horrible w/ networking and neutron | 03:23 |
| mattt | but i went on the assumption that vxlan == 1450, flat/vlan == 1500 | 03:23 |
| *** williamcaban has quit IRC | 03:24 | |
| mattt | how do network_type and physical_network fields differ? | 03:24 |
| *** Guest55355 is now known as palendae | 03:24 | |
| mattt | ah, the physical_network is the name of the physical network | 03:25 |
| mattt | why not use network_type instead ? | 03:25 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Update the neutron MTU settings on upgrade https://review.openstack.org/384635 | 03:26 |
| cloudnull | give that a spin | 03:27 |
| cloudnull | i used physical_network because while we name them to be the same thing as the type a deploy may have changed that up | 03:27 |
| irtermite | nice | 03:27 |
| cloudnull | and they may have different MTUs set for those types | 03:27 |
| mattt | yeah it'd seem to make more sense to use network_type, since that's more accurate as to what it actually is | 03:28 |
| mattt | let me re-test that | 03:28 |
| cloudnull | except in the case where my type is vlan and I set the mtu to 9000 for only 1 of the vlan nets i have | 03:29 |
| cloudnull | I guess we could pull the type and use that as a fall back when the name is unset | 03:30 |
| mattt | yeah tbh you'd know more about this than i | 03:30 |
| mattt | so perhaps use whatever method you think is best | 03:31 |
| mattt | (and confirmed, that worked) | 03:31 |
| mattt | there is one last change i'd make | 03:31 |
| mattt | i'll suggest it in the review itself | 03:31 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Update the neutron MTU settings on upgrade https://review.openstack.org/384635 | 03:34 |
| cloudnull | mattt: ^ updated to use type when going into fall back | 03:35 |
| cloudnull | so it'll use the name first and if nothing is found it'll use the type | 03:35 |
| mattt | cloudnull: cool, i have to jet but i'll try to decipher that template a bit more tomorrow :) | 03:36 |
| mattt | but seems to be working in my very limited AIO instance | 03:36 |
| *** jamielennox|away is now known as jamielennox | 03:38 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Update the neutron MTU settings on upgrade https://review.openstack.org/384635 | 03:49 |
| cloudnull | mattt: ^ implemented your sugestion on looking for mtu=0 | 03:49 |
| cloudnull | ok im off. | 03:50 |
| cloudnull | cheers all | 03:50 |
| *** rmelero has quit IRC | 03:51 | |
| *** haasn has quit IRC | 04:00 | |
| *** haasn has joined #openstack-ansible | 04:03 | |
| *** Jeffrey4l_ has quit IRC | 04:05 | |
| *** palendae has quit IRC | 04:11 | |
| *** palendae has joined #openstack-ansible | 04:15 | |
| *** palendae is now known as Guest80030 | 04:15 | |
| *** chris_hultin has quit IRC | 04:24 | |
| *** dragonmaster has quit IRC | 04:26 | |
| *** kjw3 has joined #openstack-ansible | 04:26 | |
| *** Jeffrey4l has joined #openstack-ansible | 04:26 | |
| *** dragonmaster has joined #openstack-ansible | 04:27 | |
| *** chris_hultin|AWA has joined #openstack-ansible | 04:28 | |
| *** chris_hultin|AWA is now known as chris_hultin | 04:28 | |
| *** rmelero has joined #openstack-ansible | 04:30 | |
| *** vishwanathj has quit IRC | 04:32 | |
| *** Guest80030 is now known as palendae | 04:40 | |
| *** agireud has quit IRC | 04:42 | |
| *** rmelero has quit IRC | 04:43 | |
| *** dragonmaster has quit IRC | 04:43 | |
| *** kjw3 has quit IRC | 04:43 | |
| *** dragonmaster has joined #openstack-ansible | 04:44 | |
| *** dragonmaster has quit IRC | 04:45 | |
| *** agireud has joined #openstack-ansible | 04:51 | |
| *** rmelero has joined #openstack-ansible | 04:54 | |
| *** poopcat has quit IRC | 04:55 | |
| openstackgerrit | Tuan Luong-Anh proposed openstack/openstack-ansible: Remove xrange for run both Python 2 and Python 3 https://review.openstack.org/385271 | 04:55 |
| *** agireud has quit IRC | 04:56 | |
| *** dragonmaster has joined #openstack-ansible | 04:58 | |
| *** dragonmaster has quit IRC | 04:58 | |
| *** dragonmaster has joined #openstack-ansible | 05:01 | |
| openstackgerrit | Praveen N proposed openstack/openstack-ansible: Add network config example for test and prod https://review.openstack.org/385272 | 05:02 |
| *** Jeffrey4l has quit IRC | 05:09 | |
| *** chhavi has joined #openstack-ansible | 05:11 | |
| *** Jeffrey4l has joined #openstack-ansible | 05:11 | |
| *** Jeffrey4l has quit IRC | 05:12 | |
| *** Jeffrey4l has joined #openstack-ansible | 05:13 | |
| *** Jeffrey4l has quit IRC | 05:13 | |
| *** Jeffrey4l_ has joined #openstack-ansible | 05:13 | |
| *** aetaric has quit IRC | 05:16 | |
| *** dragonmaster has quit IRC | 05:16 | |
| *** rmelero has quit IRC | 05:16 | |
| *** admin0 has joined #openstack-ansible | 05:18 | |
| *** rmelero has joined #openstack-ansible | 05:18 | |
| *** aetaric has joined #openstack-ansible | 05:25 | |
| *** Jack_Iv has joined #openstack-ansible | 05:25 | |
| *** chris_hultin has quit IRC | 05:27 | |
| *** chris_hultin|AWA has joined #openstack-ansible | 05:30 | |
| *** chris_hultin|AWA is now known as chris_hultin | 05:30 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Fix value for openstack_host_manage_hosts_file https://review.openstack.org/385277 | 05:31 |
| *** janki has joined #openstack-ansible | 05:32 | |
| *** pjm6_ has joined #openstack-ansible | 05:35 | |
| *** kylek3h_ has joined #openstack-ansible | 05:36 | |
| *** kylek3h has quit IRC | 05:38 | |
| *** pjm6 has quit IRC | 05:39 | |
| openstackgerrit | Tuan Luong-Anh proposed openstack/openstack-ansible-plugins: Remove xrange for run both Python 2 and Python 3 https://review.openstack.org/385280 | 05:39 |
| *** Jeffrey4l_ has quit IRC | 05:41 | |
| *** admin0 has quit IRC | 05:47 | |
| *** Jeffrey4l_ has joined #openstack-ansible | 05:52 | |
| *** Jeffrey4l_ has quit IRC | 05:57 | |
| *** Jeffrey4l_ has joined #openstack-ansible | 05:58 | |
| *** Jeffrey4l_ has quit IRC | 06:06 | |
| *** rmelero has quit IRC | 06:15 | |
| *** Jeffrey4l_ has joined #openstack-ansible | 06:19 | |
| *** rmelero has joined #openstack-ansible | 06:20 | |
| ioni | morning | 06:20 |
| *** rmelero has quit IRC | 06:20 | |
| *** Jeffrey4l_ has quit IRC | 06:23 | |
| *** Jeffrey4l_ has joined #openstack-ansible | 06:23 | |
| *** cristicalin has joined #openstack-ansible | 06:38 | |
| *** messy has quit IRC | 06:39 | |
| *** admin0 has joined #openstack-ansible | 06:43 | |
| *** messy has joined #openstack-ansible | 06:50 | |
| *** haad1 has joined #openstack-ansible | 06:53 | |
| *** chhavi has quit IRC | 07:00 | |
| *** chris_hultin has quit IRC | 07:10 | |
| *** cristicalin has quit IRC | 07:11 | |
| *** chris_hultin|AWA has joined #openstack-ansible | 07:12 | |
| *** chris_hultin|AWA is now known as chris_hultin | 07:12 | |
| *** asettle has joined #openstack-ansible | 07:13 | |
| *** haad1 has quit IRC | 07:13 | |
| *** asettle has quit IRC | 07:13 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova: Fix ironic compute_driver https://review.openstack.org/385308 | 07:16 |
| *** dragonmaster has joined #openstack-ansible | 07:19 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests: Fix glance_api_servers variable https://review.openstack.org/385312 | 07:20 |
| *** sdake has quit IRC | 07:22 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: Changed the home-page link https://review.openstack.org/382220 | 07:23 |
| *** rgogunskiy has joined #openstack-ansible | 07:26 | |
| *** haad1 has joined #openstack-ansible | 07:26 | |
| *** fops_ has joined #openstack-ansible | 07:27 | |
| *** williamcaban has joined #openstack-ansible | 07:28 | |
| *** fops has quit IRC | 07:30 | |
| *** javeriak has quit IRC | 07:38 | |
| *** jcrst has joined #openstack-ansible | 07:38 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Add network config example for test and prod https://review.openstack.org/385272 | 07:40 |
| *** drifterza has joined #openstack-ansible | 07:42 | |
| *** williamcaban has quit IRC | 07:45 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for Ocata 2016-10-08 https://review.openstack.org/380325 | 07:45 |
| *** fops_ has quit IRC | 07:46 | |
| *** jcrst has quit IRC | 07:46 | |
| drifterza | ehlo all | 07:46 |
| andymccr | morning drifterza! | 07:48 |
| drifterza | hows things andymccr | 07:49 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_ironic: [WIP] Enable virtualBMC testing https://review.openstack.org/381049 | 07:49 |
| andymccr | pretty good - yourself? | 07:50 |
| drifterza | agg okay, drawing diagrams. The bane of life. | 07:50 |
| andymccr | hahahahahaha | 07:51 |
| andymccr | people love pictures | 07:51 |
| drifterza | people only understand pictures dude. | 07:51 |
| evrardjp | good morning | 07:52 |
| andymccr | mornin' evrardjp! | 07:55 |
| evrardjp | fun spec in neutron coming out https://review.openstack.org/#/c/360620/6/specs/ocata/bgp-evpn-advertisment.rst | 07:56 |
| evrardjp | andymccr: did you pass your coa? | 07:56 |
| andymccr | evrardjp: mine is tomorrow | 07:57 |
| andymccr | so not yet :) | 07:57 |
| evrardjp | k | 07:57 |
| *** ssbarnea has joined #openstack-ansible | 07:59 | |
| ioni | guys, i have questions regarding haproxy and the new ssl thingy | 08:06 |
| ioni | how can one make that use a valid ssl and also be valid on an ip? | 08:06 |
| ioni | ssls are valid on domains | 08:06 |
| ioni | unless i can configure it in openstack_user_config to use domain instead of ip | 08:07 |
| ioni | as a feedback from M->N, i noticed that old configs are not purged | 08:08 |
| ioni | -rw-r--r-- 1 root root 1.4K Oct 11 22:48 ceilometer-agent-notification.conf | 08:09 |
| ioni | -rw-r--r-- 1 root root 1.2K Jul 20 15:51 ceilometer-api.conf | 08:09 |
| ioni | -rw-r--r-- 1 root root 1.3K Oct 11 22:48 ceilometer-polling.conf | 08:09 |
| ioni | ceilometer-api.conf should have been purged, because the api now is in wsgi apache | 08:09 |
| ioni | same with haproxy, there were nova console configuration duplicate, old and new | 08:09 |
| ioni | to be sure that i'm not running old services, should i destroy the containers and recreate them? | 08:10 |
| evrardjp | andymccr: odyssey4me do you have prio reviews for newton right now? | 08:11 |
| evrardjp | ioni you can use a name in the external lb part of your openstack_user_config | 08:11 |
| ioni | evrardjp, ok | 08:13 |
| evrardjp | ioni I'm not sure about your configuration duplicates, everything should be reconfigured for the latest version, what do you mean? | 08:13 |
| evrardjp | ioni pay attention, changing this cause a lot of issues | 08:13 |
| evrardjp | because you will change your openstack endpoints | 08:13 |
| evrardjp | so depending on what you need, you may think twice | 08:13 |
| ioni | evrardjp, about the external lb part? | 08:13 |
| evrardjp | yes | 08:14 |
| ioni | i really liked the old way without having everything through a global ssl | 08:14 |
| ioni | i'll look to see if i can disable it | 08:14 |
| evrardjp | oh I understand better what you mean now | 08:14 |
| evrardjp | You'd prefer going back to http by default instead of https | 08:15 |
| ioni | yes | 08:15 |
| openstackgerrit | Merged openstack/openstack-ansible: Use UCA for keepalived by default https://review.openstack.org/385148 | 08:15 |
| evrardjp | well at that time I was fighting to have https myself, so I guess we have different needs :p | 08:15 |
| ioni | https even if i configure it with a valid ssl, it won't be valid because isn't configured on an ip | 08:15 |
| ioni | not on a domain | 08:15 |
| odyssey4me | andymccr FYI https://review.openstack.org/385334 - adding a skip for all os_ repositories for CentOS7 where they don't have it in newton yet | 08:16 |
| evrardjp | I understand it's hard to move when you've defined your endpoints and work patterns already | 08:16 |
| andymccr | odyssey4me: nice! | 08:17 |
| odyssey4me | evrardjp yes, priority reviews relating to the newton release are https://review.openstack.org/#/q/starredby:jesse-pretorius+project:%255Eopenstack/openstack-ansible.*+status:open | 08:17 |
| ioni | evrardjp, about the old configs | 08:17 |
| ioni | lets take as example my ceilomenter api container | 08:17 |
| evrardjp | ok odyssey4me | 08:18 |
| ioni | https://paste.xinu.at/E8S7/ | 08:18 |
| ioni | you see there 13.2.0/bin/ceilometer-api has been started? | 08:18 |
| ioni | because there is still a /etc/init/ceilometer-api.conf | 08:18 |
| evrardjp | yes, I see | 08:19 |
| ioni | the old config was not purged | 08:19 |
| evrardjp | let me have a look it's been a while I didn't check this | 08:19 |
| ioni | same thing happen with haproxy configs | 08:19 |
| ioni | -rw-r--r-- 1 root root 390 Jun 28 12:11 ironic_api | 08:20 |
| ioni | -rw-r--r-- 1 root root 1.2K Oct 11 20:31 keystone_admin | 08:20 |
| ioni | -rw-r--r-- 1 root root 589 Jun 28 12:11 keystone_internal | 08:20 |
| ioni | -rw-r--r-- 1 root root 1.1K Oct 11 20:31 keystone_service | 08:20 |
| ioni | the keystone internal should have been purged | 08:20 |
| evrardjp | let's take the two cases independently | 08:21 |
| ioni | right | 08:21 |
| evrardjp | first the haproxy because it's easier for me :D | 08:21 |
| evrardjp | you're listing conf.d right? | 08:21 |
| ioni | yes | 08:21 |
| evrardjp | and everything in your conf.d appears in your haproxy.cfg now | 08:21 |
| evrardjp | so the upgrade should have removed what's not deployed in conf.d | 08:22 |
| evrardjp | anymore* | 08:22 |
| odyssey4me | that'll need a bug which we can reference | 08:22 |
| evrardjp | so you're right there could be a bug there, due to the way we are generating our haproxy config | 08:22 |
| evrardjp | that deserves a bug :D | 08:22 |
| odyssey4me | I think that ceilometer's API service now *has* to run behind a web server, so the clean up of old conf files and init files needs to happen in upgrades too | 08:23 |
| odyssey4me | that should be another bug | 08:23 |
| evrardjp | odyssey4me: oh ok I wasn't aware of this | 08:23 |
| evrardjp | ioni: could you file the bugs please? | 08:23 |
| evrardjp | please mention it's for upgrades :p | 08:23 |
| odyssey4me | evrardjp I think - I vaguely remember a last minute change for ceilometer or something in telemetry. | 08:23 |
| odyssey4me | ioni yes, please add 'M>N Upgrade:' as a prefix to the subjet line of your bug | 08:24 |
| evrardjp | well it makes sense to remove init file as ioni said | 08:24 |
| ioni | evrardjp, about haproxy.cfg, i cannot say, i didn't saved it | 08:24 |
| evrardjp | ioni: what do you mean with your haproxy.cfg? | 08:25 |
| ioni | <evrardjp> and everything in your conf.d appears in your haproxy.cfg now | 08:25 |
| ioni | <evrardjp> so the upgrade should have removed what's not deployed in conf.d | 08:25 |
| evrardjp | It would be important for the haproxy bug to dump your configuration | 08:25 |
| ioni | i cannot be sure about that one | 08:25 |
| ioni | i only saved conf.d as conf.d.old | 08:25 |
| ioni | i was thinking that it takes the configuration from there, like it was in the past | 08:26 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh: Update paste, policy and rootwrap configurations 2016-10-12 https://review.openstack.org/385339 | 08:26 |
| ioni | now i see that whatever is in that folder is now in haproxy.cfg | 08:26 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: Update paste, policy and rootwrap configurations 2016-10-12 https://review.openstack.org/385340 | 08:26 |
| evrardjp | ioni: haproxy role now uses conf.d to generate one big haproxy.cfg due to the way systemd works by default | 08:26 |
| evrardjp | and haproxy* | 08:26 |
| ioni | right | 08:26 |
| ioni | i have to take in consideration to edit haproxy.cfg when debugging and not conf.d :) | 08:27 |
| evrardjp | so if there is a conf.d file that shouldn't be there, it will still be there for generating the haproxy.cfg | 08:27 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat: Update paste, policy and rootwrap configurations 2016-10-12 https://review.openstack.org/385341 | 08:27 |
| evrardjp | ioni the problem with haproxy is double | 08:27 |
| evrardjp | or could be double | 08:27 |
| ioni | ok, i'll submit both bugs | 08:28 |
| evrardjp | for haproxy: I'd like to know which init of haproxy you have, and what happens if you completely flush the conf.d folder and restart haproxy manually | 08:28 |
| evrardjp | well not completely flush, keep one for fun | 08:28 |
| ioni | i have ubuntu 14 sa it's upstrat init | 08:29 |
| ioni | upstart | 08:29 |
| evrardjp | oh there is even a better solution to know this | 08:29 |
| evrardjp | could you simply do a ps aux | grep haproxy in the bug? | 08:29 |
| evrardjp | and drop the haproxy.cfg | 08:30 |
| evrardjp | that would be good enough for me | 08:30 |
| ioni | is just using the global haxproy.cfg | 08:30 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum: Update paste, policy and rootwrap configurations 2016-10-12 https://review.openstack.org/385343 | 08:30 |
| evrardjp | ok | 08:30 |
| evrardjp | so the init upgrade is fine, it's just remnants in conf.d | 08:30 |
| ioni | yes | 08:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for Ocata 2016-10-12 https://review.openstack.org/380325 | 08:31 |
| *** berendt has joined #openstack-ansible | 08:32 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: Set default keepalived cidr if none is provided https://review.openstack.org/385069 | 08:32 |
| ioni | https://bugs.launchpad.net/openstack-ansible/+bug/1632604 | 08:33 |
| openstack | Launchpad bug 1632604 in openstack-ansible "During upgrare M->N old ceilometer startap configurations are not purged" [Undecided,New] | 08:33 |
| ioni | first one | 08:33 |
| ioni | https://bugs.launchpad.net/openstack-ansible/+bug/1632609 | 08:36 |
| openstack | Launchpad bug 1632609 in openstack-ansible "During upgrare M->N old haproxy configurations from conf.d are not purged" [Undecided,New] | 08:36 |
| ioni | second one | 08:36 |
| odyssey4me | evrardjp I wonder whether https://review.openstack.org/385069 should include a release note with something like a 'known issue' describing that if the right var is not provided, keepalived will default to that CIDR? | 08:38 |
| ioni | about ceilomenter and ssl(rabbitmq) is there a way to resolve this problem? https://paste.xinu.at/32K2N/ | 08:38 |
| ioni | :SSL routines:SSL3_WRITE_PENDING:bad write retry | 08:38 |
| ioni | apart from disable ssl communication with rabbitmq | 08:39 |
| evrardjp | odyssey4me: we include in the docs that it should be defined | 08:39 |
| evrardjp | must be | 08:39 |
| odyssey4me | evrardjp yeah, I just figured that maybe we can highlight it in the release notes that if you get that CIDR then it shows that you haven't set that var | 08:39 |
| evrardjp | I'd be enclined to rewrite all these vars into group vars, and avoid the vars_files | 08:40 |
| evrardjp | so release note as known issue for newton and fix it in ocata? | 08:41 |
| odyssey4me | andymccr backport review please: https://review.openstack.org/385277 | 08:41 |
| odyssey4me | evrardjp known issue for newton, yes - and fix in ocata if there's time | 08:42 |
| *** williamcaban has joined #openstack-ansible | 08:42 | |
| odyssey4me | ie add the release note into that patch, we'll backport it... then go from there | 08:42 |
| evrardjp | thanks captain obvious ! | 08:42 |
| evrardjp | :p | 08:42 |
| odyssey4me | :p | 08:42 |
| odyssey4me | clearly someone's feeling better | 08:43 |
| evrardjp | well I can almost open both eyes at the same time for 5 minutes now. | 08:44 |
| *** williamcaban has quit IRC | 08:46 | |
| odyssey4me | andymccr evrardjp there are some backports here which need review please https://review.openstack.org/#/q/branch:stable/newton+project:%255Eopenstack/openstack-ansible.*+status:open | 08:47 |
| odyssey4me | evrardjp if I have a dict list, and for each item in the list I want to validate whether a file exists, then action a follow-on task based on the result - how do I do that? | 08:52 |
| odyssey4me | the problem here is that each tempest plugin we install might also need the packages in test-requirements.txt installed - but not every repository will necessarily have test-requirements.txt | 08:53 |
| odyssey4me | so we have to only try to install those if the file exists | 08:53 |
| evrardjp | does that work with content? | 08:53 |
| evrardjp | I'd say what you are probably looking for is (omit) | 08:53 |
| odyssey4me | so you're saying use a lookup with omit? | 08:54 |
| evrardjp | that's what I've thought at first sight, but here I'd say more simply stat register with the dict items, and then another task to do you action based on the registered result | 08:55 |
| *** oneswig has joined #openstack-ansible | 08:55 | |
| odyssey4me | can you capture the results of a stat for each file in its own register? | 08:56 |
| odyssey4me | it the stat task is acting on a list I mean | 08:56 |
| *** evilt0ne has quit IRC | 08:58 | |
| odyssey4me | evrardjp heh, apparently you can: http://www.it-recipes.com/articles/blog/206 | 08:59 |
| odyssey4me | it seems that when you use register with a loop, it keeps the results in a 'results' list which you can iterate over | 08:59 |
| evrardjp | yes I know :) | 09:00 |
| evrardjp | I think you have examples already in osa | 09:00 |
| odyssey4me | well, I didn't :p | 09:00 |
| evrardjp | https://github.com/openstack/openstack-ansible/blob/stable/mitaka/scripts/upgrade-utilities/playbooks/rfc1034_1035-cleanup.yml#L92-L97 | 09:01 |
| evrardjp | just an example | 09:02 |
| *** fops has joined #openstack-ansible | 09:05 | |
| *** fops has joined #openstack-ansible | 09:05 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Cater for situation where tempest plugins have no test-requirements.txt https://review.openstack.org/385355 | 09:06 |
| odyssey4me | evrardjp could you please take a look through https://review.openstack.org/385272 ? | 09:09 |
| evrardjp | yup just a sec | 09:10 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova: Ensure that novnc/spice consoles use the public endpoint protocol https://review.openstack.org/385011 | 09:11 |
| ioni | nice | 09:13 |
| ioni | ^ | 09:13 |
| ioni | that one i really need | 09:13 |
| ioni | my novnc didn't work and couldn't figure it out yet why | 09:13 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: Set default keepalived cidr if none is provided https://review.openstack.org/385069 | 09:14 |
| *** chhavi has joined #openstack-ansible | 09:16 | |
| *** chris_hultin has quit IRC | 09:26 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for Ocata 2016-10-12 https://review.openstack.org/380325 | 09:27 |
| *** chris_hultin|AWA has joined #openstack-ansible | 09:28 | |
| *** fops has quit IRC | 09:28 | |
| ioni | i noticed that now the number of endpoint after upgrade is greater | 09:28 |
| *** chris_hultin|AWA is now known as chris_hultin | 09:28 | |
| ioni | compared with a newly deployed N | 09:28 |
| ioni | for example, the public url is twice, once with http and once with https | 09:28 |
| *** fops has joined #openstack-ansible | 09:28 | |
| ioni | on a newly deployed N the public url is only on https | 09:28 |
| odyssey4me | ioni ah, so that's another bug where we need to clean things up | 09:29 |
| ioni | i'm rocking ! | 09:29 |
| ioni | finding bugs all over the place | 09:29 |
| ioni | maybe it's intentionally | 09:29 |
| ioni | for back compatibility? | 09:30 |
| *** haad1 has quit IRC | 09:31 | |
| *** asettle has joined #openstack-ansible | 09:36 | |
| odyssey4me | evrardjp do you mind if I tweak https://review.openstack.org/#/c/385069/4/releasenotes/notes/keepalived-default-cidr-workaround-8f2b5a0b074898e1.yaml slightly? | 09:39 |
| odyssey4me | ioni it's not intentional | 09:39 |
| ioni | odyssey4me, ok | 09:39 |
| ioni | odyssey4me, can i pm you a bit? | 09:39 |
| odyssey4me | ioni if need be - discussions are generally more beneficial for everyone in the channel | 09:40 |
| *** haad1 has joined #openstack-ansible | 09:40 | |
| ioni | i want to resolve that public ssl thingy and somebody mentioned that it is risky because endpoints are changing(which is ofcourse logically) | 09:40 |
| ioni | i used as haproxy public an IP | 09:40 |
| ioni | now any ssl i'm going to configure, it's not going to be valid | 09:41 |
| ioni | did somebody tried to change the endpoints using ansible? | 09:41 |
| ioni | should i try to do it manually from db and then re-run the playbooks? | 09:41 |
| *** pjm6_ is now known as pjm6 | 09:46 | |
| pjm6 | good morning all | 09:46 |
| *** pjm6 has quit IRC | 09:47 | |
| openstackgerrit | Merged openstack/openstack-ansible: Fix value for openstack_host_manage_hosts_file https://review.openstack.org/385277 | 09:48 |
| *** palendae has quit IRC | 09:49 | |
| *** oneswig has quit IRC | 09:51 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update role SHAs and package pins for 14.0.0 2016-10-12 https://review.openstack.org/385370 | 09:52 |
| *** palendae has joined #openstack-ansible | 09:52 | |
| *** palendae is now known as Guest92927 | 09:53 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update role SHAs for 14.0.0 2016-10-12 https://review.openstack.org/385370 | 09:53 |
| *** pjm6 has joined #openstack-ansible | 09:55 | |
| *** chhavi has quit IRC | 09:57 | |
| ioni | odyssey4me, done. https://bugs.launchpad.net/openstack-ansible/+bug/1632645 | 09:58 |
| openstack | Launchpad bug 1632645 in openstack-ansible "During upgrare M->N old endpoints are not cleaned" [Undecided,New] | 09:58 |
| odyssey4me | thanks ioni | 09:58 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova: Fix ironic compute_driver https://review.openstack.org/385308 | 10:00 |
| *** admin0 has quit IRC | 10:02 | |
| *** chhavi has joined #openstack-ansible | 10:05 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests: Fix glance_api_servers variable https://review.openstack.org/385312 | 10:06 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Set default keepalived cidr if none is provided https://review.openstack.org/385069 | 10:10 |
| *** asettle has quit IRC | 10:17 | |
| *** haad1 has quit IRC | 10:18 | |
| *** Mudpuppy has joined #openstack-ansible | 10:22 | |
| *** haad1 has joined #openstack-ansible | 10:23 | |
| openstackgerrit | Praveen N proposed openstack/openstack-ansible: [docs] Add network config example for test and prod https://review.openstack.org/385272 | 10:23 |
| *** Praveen has joined #openstack-ansible | 10:28 | |
| *** asettle has joined #openstack-ansible | 10:29 | |
| Praveen | Hi | 10:31 |
| Praveen | Hello Jesse, please provide suggestions for bug 1631321 | 10:33 |
| openstack | bug 1631321 in openstack-ansible "Docs: Test deployment shouldn't talk about bonds" [High,In progress] https://launchpad.net/bugs/1631321 - Assigned to Praveen N (praveenn) | 10:33 |
| openstackgerrit | Ravi Gummadi proposed openstack/openstack-ansible-repo_build: Add cython to repo_pip_packages https://review.openstack.org/385388 | 10:34 |
| *** janki has quit IRC | 10:36 | |
| *** Praveen__ has joined #openstack-ansible | 10:39 | |
| *** Praveen__ has quit IRC | 10:40 | |
| *** Praveen__ has joined #openstack-ansible | 10:42 | |
| *** Mudpuppy has quit IRC | 10:45 | |
| *** Praveen__ has quit IRC | 10:46 | |
| odyssey4me | Hi Praveen - thanks for joining us here! I have no specific feedback at the moment - perhaps evrardjp will. It's ideal if you're able to be in the channel to field any questions or discuss options available. | 10:52 |
| Praveen | sure, thanks a lot | 10:53 |
| Praveen | I've modified openstack_interface.cfg.test.example, please review & let me know for any changes | 10:54 |
| *** chhavi has quit IRC | 10:57 | |
| *** ssbarnea has quit IRC | 10:57 | |
| odyssey4me | Praveen yeah, it's not my area of expertise - I'm hoping that evrardjp, cloudnull or jamesdenton will pick up the next round of reviews some time soon | 10:58 |
| Praveen | ok :) | 10:59 |
| odyssey4me | andymccr heh, cute - check the way that the rally results show up in the gate: http://logs.openstack.org/60/385360/1/check/gate-rally-dsvm-neutron-rally/0a0614d/ | 11:05 |
| odyssey4me | I've seen something similar for devstack | 11:05 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Cater for situation where tempest plugins have no test-requirements.txt https://review.openstack.org/385355 | 11:12 |
| *** spotz is now known as spotz_zzz | 11:12 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Install test requirements for plugins https://review.openstack.org/384942 | 11:15 |
| andymccr | odyssey4me: fancy fancy. | 11:16 |
| odyssey4me | andymccr yeah, I like it :) | 11:17 |
| odyssey4me | we should do that | 11:17 |
| odyssey4me | we'll need a different index for the role tests and for the integrated build though | 11:17 |
| odyssey4me | but that's certainly helpful to help orientate contributors | 11:17 |
| *** RaviGummadi has joined #openstack-ansible | 11:18 | |
| odyssey4me | lemme work up a bug quickly | 11:18 |
| *** haad1 has quit IRC | 11:21 | |
| odyssey4me | done https://bugs.launchpad.net/openstack-ansible/+bug/1632670 | 11:21 |
| openstack | Launchpad bug 1632670 in openstack-ansible "Add html index to role & integrated repo results" [Wishlist,New] | 11:21 |
| odyssey4me | oh, and just for fun - it seems like we're not producing any logs for the containers | 11:23 |
| *** evilt0ne has joined #openstack-ansible | 11:27 | |
| *** evilt0ne has quit IRC | 11:27 | |
| *** evilt0ne has joined #openstack-ansible | 11:27 | |
| *** askb has quit IRC | 11:30 | |
| openstackgerrit | Deepak proposed openstack/openstack-ansible: [Docs] Fix the listing of ops-remove-computehost.rst https://review.openstack.org/385406 | 11:36 |
| drifterza | wow | 11:40 |
| drifterza | almost the whole day on 1 diagram | 11:40 |
| drifterza | kill me | 11:41 |
| ioni | evrardjp, i'm starting to like haproxy ssl now :D | 11:42 |
| ioni | works like a charm :) | 11:42 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_ironic: [WIP] Enable virtualBMC testing https://review.openstack.org/381049 | 11:43 |
| odyssey4me | drifterza diagrams are hard, but very valuable when they're done right | 11:43 |
| logan- | hey evrardjp, you might be interested in this https://github.com/Logan2211/ansible-unbound/commit/a1f526df780d81685d635cabb3ab6fff2d81d804 -- i updated the tox config in my roles to run linter using latest stable ansible, and multiple envs for functional tests using different ansible versions. | 11:44 |
| logan- | I noticed the latest ansible-lint gives this error against all of our "Check init system" tasks https://travis-ci.org/Logan2211/ansible-unbound/jobs/166919724#L273 | 11:44 |
| logan- | they will probably all need a "changed_when: false" at some point | 11:45 |
| *** retreved has joined #openstack-ansible | 11:45 | |
| odyssey4me | logan- oh neat | 11:45 |
| odyssey4me | I think when we release newton we should bump our ansible & ansible-lint requirements, then deal with the consequences | 11:46 |
| *** thorst has joined #openstack-ansible | 11:47 | |
| *** haad1 has joined #openstack-ansible | 11:47 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_ironic: [WIP] Enable virtualBMC testing https://review.openstack.org/381049 | 11:48 |
| odyssey4me | logan- any chance that you could tag a release for https://github.com/logan2211/ansible-bird ? | 11:49 |
| logan- | err yes I must have missed that one | 11:49 |
| logan- | done | 11:50 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update role SHAs for 14.0.0 2016-10-12 https://review.openstack.org/385370 | 11:51 |
| odyssey4me | thanks logan- :) it looks like your fixes are all good | 11:51 |
| odyssey4me | maybe I should mark the SHA bump as 'closes-bug' ? | 11:52 |
| *** RaviGummadi has quit IRC | 11:53 | |
| *** johnmilton has joined #openstack-ansible | 11:53 | |
| *** johnmilton has joined #openstack-ansible | 11:54 | |
| odyssey4me | yeah, lemme do that | 11:54 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update role SHAs for 14.0.0 2016-10-12 https://review.openstack.org/385370 | 11:54 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-plugins: Add get_networks command to neutron library https://review.openstack.org/385423 | 11:56 |
| *** aetaric has quit IRC | 11:57 | |
| openstackgerrit | Deepak proposed openstack/openstack-ansible: [Docs] Removed extra grave accent(`) https://review.openstack.org/385425 | 11:58 |
| *** aetaric has joined #openstack-ansible | 11:58 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update role SHAs for 14.0.0 2016-10-12 https://review.openstack.org/385370 | 12:00 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-plugins: Add get_networks command to neutron library https://review.openstack.org/385423 | 12:04 |
| *** chris_hultin has quit IRC | 12:04 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [Docs] Removed extra grave accent(`) https://review.openstack.org/385425 | 12:05 |
| *** chris_hultin|AWA has joined #openstack-ansible | 12:08 | |
| *** chris_hultin|AWA is now known as chris_hultin | 12:09 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_ironic: [WIP] Enable virtualBMC testing https://review.openstack.org/381049 | 12:13 |
| *** chhavi has joined #openstack-ansible | 12:14 | |
| openstackgerrit | Merged openstack/openstack-ansible: [Docs] Removed extra grave accent(`) https://review.openstack.org/385425 | 12:14 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [Docs] Removed extra grave accent(`) https://review.openstack.org/385434 | 12:14 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-plugins: Add get_networks command to neutron library https://review.openstack.org/385423 | 12:16 |
| odyssey4me | andymccr evrardjp I'd appreciate a review of https://review.openstack.org/385355 - it fixes a bug which is blocking some role patches for master. | 12:17 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_ironic: [WIP] Enable virtualBMC testing https://review.openstack.org/381049 | 12:19 |
| openstackgerrit | Merged openstack/openstack-ansible: [Docs] Removed extra grave accent(`) https://review.openstack.org/385434 | 12:21 |
| mhayden | mornin' | 12:21 |
| *** sdake has joined #openstack-ansible | 12:21 | |
| *** sdake has quit IRC | 12:22 | |
| *** sdake has joined #openstack-ansible | 12:23 | |
| *** woodard has joined #openstack-ansible | 12:24 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Cater for situation where tempest plugins have no test-requirements.txt https://review.openstack.org/385355 | 12:25 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Install test requirements for plugins https://review.openstack.org/384942 | 12:26 |
| odyssey4me | o/ mhayden how goes it? | 12:27 |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 12:30 |
| *** williamcaban has joined #openstack-ansible | 12:30 | |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 12:32 |
| *** dragonmaster has quit IRC | 12:40 | |
| *** jheroux has joined #openstack-ansible | 12:46 | |
| *** chhavi has quit IRC | 12:47 | |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 12:48 |
| *** chhavi has joined #openstack-ansible | 12:59 | |
| *** admin0 has joined #openstack-ansible | 12:59 | |
| ioni | can somebody see if they can replicate it? https://bugs.launchpad.net/openstack-ansible/+bug/1632702 | 13:04 |
| openstack | Launchpad bug 1632702 in openstack-ansible "During upgrare M->N the public endpoint for keystone doesn't have https" [Undecided,New] | 13:04 |
| *** klamath has joined #openstack-ansible | 13:05 | |
| *** klamath has quit IRC | 13:05 | |
| *** klamath has joined #openstack-ansible | 13:06 | |
| *** cathrichardson has joined #openstack-ansible | 13:07 | |
| *** cathrichardson has quit IRC | 13:07 | |
| *** cathrichardson has joined #openstack-ansible | 13:07 | |
| openstackgerrit | Praveen N proposed openstack/openstack-ansible: [docs] Add network config example for test and prod https://review.openstack.org/385272 | 13:07 |
| *** haad1 has quit IRC | 13:08 | |
| *** haad1 has joined #openstack-ansible | 13:09 | |
| *** gouthamr has joined #openstack-ansible | 13:09 | |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 13:15 |
| *** vishwanathj has joined #openstack-ansible | 13:15 | |
| *** jamesdenton has joined #openstack-ansible | 13:19 | |
| *** admin0 has quit IRC | 13:25 | |
| drifterza | odyssey4me, you around dude ? | 13:26 |
| odyssey4me | drifterza yep | 13:26 |
| drifterza | need advise on a role | 13:27 |
| *** woodard has quit IRC | 13:27 | |
| odyssey4me | sure | 13:27 |
| drifterza | monasca is increasingly tricky to get pinned up, would you agree that having another role called monasca_backend for things like (zookeeper,kafka,storm,influxdb) and then have the os_monasca role only contain the monasca relevent things ? | 13:28 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-plugins: Add get_networks command to neutron library https://review.openstack.org/385423 | 13:28 |
| drifterza | or should I try and pin everything in one role | 13:28 |
| *** haad1 has quit IRC | 13:29 | |
| *** phschwartz has quit IRC | 13:30 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest: Install test requirements for plugins https://review.openstack.org/384942 | 13:32 |
| odyssey4me | drifterza it'd be better to have a role for each of the dependent services (zookeeper, etc), then have *only* monasca deployed in the os_monasca role | 13:33 |
| odyssey4me | ideally use existing galaxy roles for zookeeper and such | 13:33 |
| odyssey4me | if you could figure out each of those dep roles, then you can do a patch to extract them and replace them with the dep role from galaxy - one per patch | 13:35 |
| odyssey4me | that'll make the work much easier to test, review and follow | 13:35 |
| *** phschwartz has joined #openstack-ansible | 13:35 | |
| drifterza | hmmm | 13:36 |
| drifterza | I am close to having everything in the one role | 13:37 |
| drifterza | just wondering how sustainable it would be | 13:37 |
| odyssey4me | drifterza if you really, must then go ahead - but ultimately it's more ideal if we can take advantage of the broader community for those dep services | 13:38 |
| odyssey4me | there are people who know much better than we how to install and configure them, and then those communities also keep them maintained as new versions come out | 13:38 |
| odyssey4me | we ideally just want to maintain the stuff in our domain of expertise - ie OpenStack | 13:39 |
| odyssey4me | this cuts the workload for us and makes it much easier to sustain | 13:39 |
| drifterza | okay | 13:39 |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 13:39 |
| drifterza | ill work out a way to use galaxy in this PR | 13:39 |
| *** dragonmaster has joined #openstack-ansible | 13:40 | |
| odyssey4me | if you add the specific role into ansible-role-requirements.yml, then it'll get downloaded | 13:40 |
| odyssey4me | then all you need is to add it as a dependency in the meta/main.yml file and to set the right vars which the role wants | 13:41 |
| odyssey4me | try to find roles that are clearly well tested using travis, etc - and ideally already support ubuntu & centos | 13:41 |
| *** Oku_OS is now known as Oku_OS-away | 13:41 | |
| *** Oku_OS-away is now known as Oku_OS | 13:42 | |
| drifterza | thanks for the advise dude | 13:43 |
| drifterza | think thats a better way to handle this beast | 13:43 |
| *** batkins61 has joined #openstack-ansible | 13:53 | |
| *** dragonmaster_ has joined #openstack-ansible | 13:54 | |
| batkins61 | I'm still getting the following frequently, although 'glaera_root_password' is clearly defined in user_secrets.yml | 13:54 |
| batkins61 | fatal: [stlrx2540m2-18_galera_container-1669b62a] => One or more undefined variables: 'galera_root_password' is undefined | 13:54 |
| batkins61 | Any ideas on what's causing that? | 13:54 |
| *** pcaruana has joined #openstack-ansible | 13:54 | |
| *** dragonmaster has quit IRC | 13:54 | |
| *** dragonmaster_ is now known as dragonmaster | 13:54 | |
| *** Praveen has quit IRC | 13:58 | |
| *** rmelero has joined #openstack-ansible | 13:58 | |
| openstackgerrit | Merged openstack/openstack-ansible: [DOCS] Update manual upgrade guide https://review.openstack.org/385150 | 13:59 |
| *** phalmos_ has quit IRC | 14:00 | |
| *** agrebennikov has joined #openstack-ansible | 14:00 | |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 14:01 |
| *** spotz_zzz is now known as spotz | 14:03 | |
| odyssey4me | batkins61 I would guess that you're not using the 'openstack-ansible' command, but are instead using 'ansible-playbook' | 14:04 |
| batkins61 | Hmm, let me confirm that. | 14:05 |
| *** rmelero has quit IRC | 14:06 | |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client: Add newline to ceph.conf created from ceph_conf_file https://review.openstack.org/385118 | 14:07 |
| *** woodard has joined #openstack-ansible | 14:08 | |
| *** woodard has quit IRC | 14:08 | |
| batkins61 | Yep, looks like that was it according to command history. I'll try again with openstack-ansible. | 14:08 |
| *** woodard has joined #openstack-ansible | 14:09 | |
| *** haad1 has joined #openstack-ansible | 14:09 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest: Cater for situation where tempest plugins have no test-requirements.txt https://review.openstack.org/385355 | 14:10 |
| openstackgerrit | Merged openstack/openstack-ansible-repo_build: Add cython to repo_pip_packages https://review.openstack.org/385388 | 14:10 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [WIP] Provide example configurations https://review.openstack.org/385496 | 14:15 |
| *** chris_hultin has quit IRC | 14:15 | |
| odyssey4me | automagically evrardjp andymccr Can we get some reviews for https://review.openstack.org/385370 please | 14:16 |
| *** michaelgugino has joined #openstack-ansible | 14:17 | |
| *** pester has quit IRC | 14:17 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_build: Add cython to repo_pip_packages https://review.openstack.org/385498 | 14:18 |
| drifterza | odyssey4me, could you point me to why this is happening ? | 14:18 |
| drifterza | http://logs.openstack.org/55/375655/72/check/gate-openstack-ansible-os_monasca-ansible-func-ubuntu-trusty/c0e4351/console.html#_2016-10-12_14_15_24_439551 | 14:18 |
| *** chris_hultin|AWA has joined #openstack-ansible | 14:19 | |
| *** chris_hultin|AWA is now known as chris_hultin | 14:19 | |
| odyssey4me | drifterza looking | 14:19 |
| *** Mudpuppy has joined #openstack-ansible | 14:19 | |
| drifterza | Thanks | 14:19 |
| drifterza | I'm trying to use the dict group mapping lookup like in nova, but cant see why this is happening | 14:19 |
| spotz | odyssey4me: question on that review, we're pulling from private repos vs openstack ones? | 14:20 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Security: Add tasks for RHEL-07-010010 https://review.openstack.org/383943 | 14:21 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Security: Add tasks for RHEL-07-010010 https://review.openstack.org/383943 | 14:23 |
| *** Mudpuppy has quit IRC | 14:23 | |
| *** Mudpuppy has joined #openstack-ansible | 14:24 | |
| odyssey4me | drifterza yeah, I've been getting the same weird error in https://review.openstack.org/378930 and couldn't figure it out | 14:25 |
| odyssey4me | I'll comment on a few things that may help though | 14:25 |
| drifterza | okay so its not something i did wrong, cos I took the example from the nova plays and to me it should work. | 14:26 |
| *** pcaruana has quit IRC | 14:27 | |
| odyssey4me | drifterza added some comments - those adjustments *may* help | 14:28 |
| drifterza | cool testing | 14:28 |
| odyssey4me | spotz huh? | 14:29 |
| spotz | odyssey4me: ttps://github.com/willshersystems/ansible-sshd andhttps://github.com/logan2211 | 14:29 |
| *** fxpester has joined #openstack-ansible | 14:30 | |
| odyssey4me | spotz yeah, we use github as a source for repositories which are outside of OpenStack | 14:30 |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 14:31 |
| andymccr | drifterza: it looks like its failing because the group doesnt exist | 14:31 |
| spotz | odyssey4me: Ok wasn't sure if we were trying to call all the parts from openstack contralled repos | 14:31 |
| andymccr | some of those worked | 14:31 |
| drifterza | andymccr, ? | 14:31 |
| andymccr | drifterza: check the line above the one you linked - its successful, basically the "zookeeper" one is failing because the group is called "monasca_zookeeper" | 14:32 |
| andymccr | so when it evaluates groups[item.value.group] that doesnt exist so it bombs. | 14:32 |
| drifterza | ah | 14:32 |
| andymccr | all your groups are called "monasca_x" at least in the inventory for testing | 14:33 |
| drifterza | lets try that | 14:33 |
| *** rmelero has joined #openstack-ansible | 14:33 | |
| andymccr | odyssey4me: same for you and glance | 14:33 |
| andymccr | add glance_api and glance_registry to the inventory for testing | 14:33 |
| andymccr | you only have the "glance_all" group | 14:33 |
| odyssey4me | andymccr ah | 14:33 |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 14:34 |
| andymccr | so its failing | 14:34 |
| odyssey4me | <---- dumbass | 14:34 |
| drifterza | same <----> | 14:34 |
| odyssey4me | andymccr could you please comment in the review so that I can circle back to it at some point | 14:34 |
| odyssey4me | thanks for looking into it :) | 14:34 |
| *** BjoernT has joined #openstack-ansible | 14:34 | |
| andymccr | np! all good and fixed up | 14:35 |
| drifterza | lets see if it goes now | 14:40 |
| *** chhavi has quit IRC | 14:40 | |
| *** hj-hpe has joined #openstack-ansible | 14:40 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_glance: Use dictionary for service group mappings https://review.openstack.org/378930 | 14:42 |
| andymccr | any reviews on https://review.openstack.org/#/c/385423/ would be appreciated :D | 14:44 |
| *** weezS has joined #openstack-ansible | 14:49 | |
| *** rmelero has quit IRC | 14:52 | |
| *** dragonmaster has quit IRC | 14:54 | |
| drifterza | thanks andymccr | 14:55 |
| drifterza | that fixed that issue | 14:55 |
| andymccr | np! | 14:56 |
| openstackgerrit | Charles Farquhar proposed openstack/openstack-ansible-os_ironic: Use correct keystone_authtoken config keys https://review.openstack.org/385508 | 14:56 |
| drifterza | andymccr you have any experience with using program_override in the upstart/systemd files | 14:57 |
| drifterza | trying to use the dict but for certain groups I need to run a binary thats not the same as the name of the group | 14:57 |
| drifterza | monasca-api = binary = gunicorn --options | 14:58 |
| odyssey4me | andymccr I've added stevelle and hughsaunders as they're usually best (along with jmccrory and automagically) at doing python reviews | 14:58 |
| odyssey4me | cloudnull is good too - will add him now | 14:58 |
| odyssey4me | drifterza we have no precedence for that anywhere, but we discussed an option th eother day | 14:59 |
| odyssey4me | basically have another item in the dict - maybe 'service_bin' | 15:00 |
| drifterza | yup | 15:00 |
| odyssey4me | then adjust the templates to use it if it's defined | 15:00 |
| *** phalmos has joined #openstack-ansible | 15:00 | |
| openstackgerrit | Merged openstack/openstack-ansible: Set default keepalived cidr if none is provided https://review.openstack.org/385069 | 15:01 |
| odyssey4me | so instead of 'program_override' in the template, use the dict path | 15:01 |
| drifterza | hmm | 15:01 |
| odyssey4me | we can apply that updated pattern to all the other roles if that works out nicely | 15:01 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests: Move ANSIBLE_CLI_PARAMETERS after playbook path https://review.openstack.org/385511 | 15:01 |
| *** rgogunskiy has quit IRC | 15:01 | |
| drifterza | item.value.service_bin | 15:01 |
| drifterza | for example | 15:01 |
| *** rmelero has joined #openstack-ansible | 15:01 | |
| drifterza | in the template now | 15:02 |
| odyssey4me | also, I would guess that program_config_options in the template should be replaced by something in the dict like service_init_options | 15:02 |
| odyssey4me | yes, something like that | 15:02 |
| drifterza | this is just for upstart/systemd | 15:03 |
| drifterza | not specific app overrides. | 15:03 |
| odyssey4me | well, maybe service_init_bin then? | 15:03 |
| drifterza | yeah lets use that | 15:03 |
| odyssey4me | if upstart and systemd need different values, then have it be something like service_upstart_bin / service_upstart_options | 15:04 |
| drifterza | ill test it out | 15:05 |
| drifterza | thanks | 15:05 |
| *** sdake has quit IRC | 15:05 | |
| *** sdake has joined #openstack-ansible | 15:06 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Set default keepalived cidr if none is provided https://review.openstack.org/385514 | 15:06 |
| *** batkins61 has quit IRC | 15:06 | |
| *** batkins61 has joined #openstack-ansible | 15:07 | |
| *** cmart has joined #openstack-ansible | 15:14 | |
| andymccr | drifterza: i haven't but I think you just set the program_override var to be the exec name that you want. I'd imagine you would want to change it to be "item.value.program_override" and then specify it in the monasca_services dict | 15:18 |
| andymccr | i think it has to be the full path (which would be the only caveat) | 15:19 |
| *** TxGirlGeek has joined #openstack-ansible | 15:19 | |
| *** fops has quit IRC | 15:19 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [WIP][docs] Provide example configurations https://review.openstack.org/385496 | 15:19 |
| drifterza | yeah thats okay cause the gunicorn is in the venv path | 15:19 |
| drifterza | but Ill give it a test | 15:19 |
| *** jimbaker has joined #openstack-ansible | 15:20 | |
| andymccr | then probably just do like "{{monasca_bin}}/gunicorn" as your program_override | 15:20 |
| openstackgerrit | Charles Farquhar proposed openstack/openstack-ansible-os_ironic: Use correct keystone_authtoken config keys https://review.openstack.org/385523 | 15:20 |
| drifterza | yup im thinking of adding something like this to the dict: service_init_bin: "{{ monasca_bin}}/gunicorn" , service_init_options: "--listen etc.. etc.." | 15:21 |
| drifterza | then I have to mod the upstart/systemd templates to look for these | 15:21 |
| drifterza | or is this over kill ? | 15:22 |
| andymccr | that sounds ok to me - i mean the program_override bit is already there so its just about changing it to pull from the dict - which makes sense to me, i think it's there as a relic of how we used to set these up | 15:23 |
| *** timrc_ is now known as timrc | 15:24 | |
| drifterza | is it worth changing it though to have bin and options to bin | 15:24 |
| *** jamesdenton has quit IRC | 15:24 | |
| evrardjp | hey I'm back | 15:26 |
| andymccr | drifterza: i think-so. the way we're doing it with dicts now (here and in a few others) suggests to me that all the values should be in that dict | 15:26 |
| andymccr | how we used to do it is individually call the "init_common.yml" as many times as there are services with those variables set as "vars:" on the include | 15:26 |
| drifterza | so +1 for the service_init_bin/service_init_options ? | 15:27 |
| drifterza | or just use program_override | 15:27 |
| evrardjp | odyssey4me: you know I don't mind you tweaking the bits | 15:27 |
| evrardjp | as long as it stays honorable! | 15:27 |
| andymccr | drifterza: would init options be like the "program_config_options" ? | 15:28 |
| drifterza | yes | 15:28 |
| andymccr | and then system_init_bin like program_override | 15:28 |
| andymccr | sure yeah that makes sense to me | 15:28 |
| drifterza | like flags to the bin | 15:28 |
| evrardjp | I still hate this patch, but that's another story | 15:28 |
| evrardjp | Ok I can get back to my todo list... | 15:29 |
| andymccr | drifterza: yeah all good i think, the only reason program_override exists is because in most cases the service name is actually the binary name too - so its simple where that pattern matches you dont need to specify it | 15:30 |
| evrardjp | logan-: I think I'll shamelessly adapt my CI to be closer to yours | 15:30 |
| drifterza | andymccr, 100% | 15:32 |
| drifterza | its sometimes useful to add options even though the bin is the same name | 15:32 |
| evrardjp | drifterza: I still have tests with trove on my todolist, is it still current or it's too late now? | 15:33 |
| *** kylek3h_ is now known as kylek3h | 15:33 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Provide example configurations https://review.openstack.org/385496 | 15:33 |
| drifterza | there is some work for trove still, getting trove to upload images to glance, and use tempest to test functionality | 15:33 |
| drifterza | so functional tests needs to be worked on evrardjp | 15:34 |
| odyssey4me | evrardjp andymccr asettle I'd love some reviews of this docs change which provides the example configs we've all been waiting for. :p https://review.openstack.org/385496 | 15:34 |
| odyssey4me | spotz ^ | 15:35 |
| asettle | odyssey4me: if you insist | 15:35 |
| spotz | odyssey4me: Got it | 15:35 |
| drifterza | evrardjp, I guess the answer you looking for is, yes it is still current | 15:35 |
| drifterza | :D | 15:36 |
| evrardjp | ok I will have a look at everyone's demands | 15:36 |
| logan- | evrardjp: sounds good :) personally I think it will make maintaining my roles easier by running linters and tests using latest stable ansible to fix things upstream far ahead of OSA's ansible bumps. | 15:36 |
| evrardjp | I'll give answers demanding on my supply | 15:36 |
| evrardjp | logan-: I'm myself using ansible 1.9 and ansible 2. as requirements for my two types of tests | 15:37 |
| evrardjp | so I always upgrade to latest ansible | 15:37 |
| evrardjp | it's generally fine | 15:37 |
| evrardjp | but then I don't do periodic jobs, and that's terrible | 15:37 |
| *** Guest92927 is now known as palendae | 15:38 | |
| odyssey4me | evrardjp in https://review.openstack.org/385496 I took a guess at how to make haproxy/keepalived work across multiple nodes... please correct my horrible guess work :) | 15:39 |
| odyssey4me | the idea is to provide a minimal and simple set of examples to help people understand how it works | 15:40 |
| *** fguillot has quit IRC | 15:40 | |
| evrardjp | odyssey4me: is that the proper link? | 15:42 |
| *** rmelero has quit IRC | 15:42 | |
| *** admin0 has joined #openstack-ansible | 15:42 | |
| evrardjp | lol @1.2.3.4/32 | 15:42 |
| evrardjp | :D | 15:42 |
| odyssey4me | evrardjp the one thing I'm not clear on is how keepalived actually sets the vip for the multinode environment | 15:43 |
| *** rmelero has joined #openstack-ansible | 15:43 | |
| evrardjp | what do you mean? | 15:44 |
| evrardjp | it's using vrrp | 15:44 |
| asettle | odyssey4me: all yours | 15:44 |
| evrardjp | let me review this :D | 15:44 |
| odyssey4me | well, at some point it actually needs to have an IP - in the docs I set one, but in the config I don't know where to assign it so that it binds it | 15:44 |
| spotz | heheh I'll wait for asettle's changes to get in:) | 15:45 |
| *** david-lyle has joined #openstack-ansible | 15:46 | |
| asettle | Two eyes are always better than one :D that'd be great if you can review after spotz :) | 15:46 |
| evrardjp | odyssey4me: I thought at some point to do a {{ lookup('dns',external_lb_vip_address) }} | 15:46 |
| evrardjp | but it's not a terribly good idea | 15:46 |
| odyssey4me | lemme change based on asettle's feedback quickly | 15:47 |
| spotz | asettle Yeah I figured I'd give him a chance to make yours just in case they end up changing things | 15:47 |
| evrardjp | then you have to guess the cidr for the nic, which is fine based on facts, but it could be cleaner | 15:47 |
| asettle | spotz: excellente :) thanks odyssey4me | 15:48 |
| *** rmelero has quit IRC | 15:48 | |
| asettle | I'll be off shortly, I was in early as. But i'll review again after spotz :) (we'll go round and round and round ) | 15:48 |
| spotz | heheh | 15:48 |
| *** david-lyle has quit IRC | 15:52 | |
| *** asettle has quit IRC | 15:53 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Check for configured groups in the environment https://review.openstack.org/383920 | 15:53 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Provide example configurations https://review.openstack.org/385496 | 15:55 |
| odyssey4me | ^ still working on it | 15:57 |
| *** adrian_otto has joined #openstack-ansible | 15:57 | |
| drifterza | odyssey4me, so I update ansible-role-requirements, the -name: zookeeper , can the name be anything I want ? | 16:01 |
| *** jamesdenton has joined #openstack-ansible | 16:01 | |
| evrardjp | that's the rolename | 16:01 |
| odyssey4me | drifterza the name should be the name used by the role in ansible galaxy | 16:01 |
| evrardjp | not really | 16:02 |
| drifterza | ah so author.name | 16:02 |
| *** cmart has quit IRC | 16:02 | |
| evrardjp | name is the name on your host | 16:02 |
| evrardjp | it could be the same | 16:02 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Provide example configurations https://review.openstack.org/385496 | 16:02 |
| evrardjp | it's not forced to be the same | 16:02 |
| drifterza | so at the moment I did, -name: grafana | 16:02 |
| drifterza | but on galaxy its : tkuhlman.monasca-ui | 16:02 |
| evrardjp | if you just do - then the item will be the galaxy name | 16:02 |
| evrardjp | you can use that in src | 16:03 |
| evrardjp | iirc | 16:03 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Provide example configurations https://review.openstack.org/385496 | 16:03 |
| odyssey4me | spotz ^ OK, ready to go | 16:03 |
| evrardjp | drifterza: http://docs.ansible.com/ansible/galaxy.html#installing-multiple-roles-from-multiple-files | 16:03 |
| spotz | odyssey4me: Okie:) | 16:04 |
| evrardjp | src is the src, if you don't give full url, it's assumed to be a galaxy role and will look there | 16:04 |
| evrardjp | else it will pack a gz from the url you give based on the scm | 16:04 |
| *** thorst is now known as thorst_afk | 16:04 | |
| drifterza | aaah | 16:05 |
| evrardjp | and will copy it into your roles_folder/<name> | 16:05 |
| drifterza | so if I put the github url | 16:05 |
| drifterza | in src | 16:05 |
| drifterza | and use scm:git | 16:05 |
| drifterza | it wont use galaxy ? | 16:05 |
| *** weezS_ has joined #openstack-ansible | 16:05 | |
| evrardjp | it won't use galaxy website and api | 16:05 |
| evrardjp | it will still use ansible-galaxy cli :p | 16:05 |
| *** david-lyle has joined #openstack-ansible | 16:06 | |
| *** weezS has quit IRC | 16:06 | |
| *** weezS_ is now known as weezS | 16:06 | |
| drifterza | so its just better to use -src | 16:06 |
| drifterza | okay | 16:06 |
| odyssey4me | mhayden is https://review.openstack.org/385511 what's screwing up the use of extra vars? | 16:08 |
| odyssey4me | I've been trying to figure that one out for two days | 16:08 |
| openstackgerrit | Praveen N proposed openstack/openstack-ansible: [docs] Add network config example for test and prod https://review.openstack.org/385272 | 16:09 |
| drifterza | would this work : http://paste.openstack.org/show/585492/ | 16:11 |
| drifterza | for my meta/main.yml | 16:11 |
| *** chhavi has joined #openstack-ansible | 16:12 | |
| drifterza | and this for my ansible-role-requirements.yml : http://paste.openstack.org/show/585493/ | 16:13 |
| drifterza | evrardjp, ? | 16:13 |
| evrardjp | sorry | 16:14 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh: Update paste, policy and rootwrap configurations 2016-10-12 https://review.openstack.org/385542 | 16:14 |
| evrardjp | yup | 16:14 |
| evrardjp | it could work | 16:14 |
| evrardjp | theoretically | 16:15 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: Update paste, policy and rootwrap configurations 2016-10-12 https://review.openstack.org/385543 | 16:15 |
| evrardjp | however in the gates we are almost never using that | 16:15 |
| evrardjp | because we now use git by default :p | 16:15 |
| drifterza | so now in my defaults/main.yml for my role, I just need to cater for those galaxy roles required variable | 16:15 |
| evrardjp | and we bypass the ansible-galaxy cli completely | 16:15 |
| drifterza | rofl | 16:15 |
| evrardjp | but theoretically it's the good usage | 16:15 |
| spotz | odyssey4me: one change | 16:15 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat: Update paste, policy and rootwrap configurations 2016-10-12 https://review.openstack.org/385544 | 16:16 |
| drifterza | blah, so how would I use external roles to this monasca role | 16:16 |
| evrardjp | anyway now galaxy is open source | 16:16 |
| *** rmelero has joined #openstack-ansible | 16:16 | |
| odyssey4me | yes, we bypass ansible-galaxy's CLI because the API is flaky for gate usage | 16:16 |
| evrardjp | we can have our own | 16:16 |
| drifterza | sigh | 16:16 |
| logan- | drifterza some revisions http://cdn.pasteraw.com/jdyen2bfxhsucdiy6o4vylidf9v2uos but also you don't have a git source listed for monasca-ui as you do the other ones | 16:16 |
| drifterza | logan-, would that work in the gate ? | 16:16 |
| evrardjp | drifterza: what I'd do if I were you, is to find the real url on every galaxy role | 16:16 |
| logan- | the role will be downloaded by osa bootstrap or ansible-galaxy using which ever name you place in ansible-role-requirements. then you can call the dependency using that name, not the full galaxy name | 16:16 |
| evrardjp | and put that into the requirements | 16:17 |
| evrardjp | and define the name you want, this way it's more explicit etc | 16:17 |
| drifterza | ah bonus | 16:17 |
| drifterza | thats what I wanted | 16:17 |
| evrardjp | yup that's the cleaner way to work in all worlds | 16:18 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum: Update paste, policy and rootwrap configurations 2016-10-12 https://review.openstack.org/385546 | 16:18 |
| *** cmart has joined #openstack-ansible | 16:18 | |
| evrardjp | it works with our clone process and it works with galaxy cli | 16:18 |
| evrardjp | but now you understand the whole story :p | 16:18 |
| drifterza | lol | 16:19 |
| drifterza | thanks guys | 16:19 |
| drifterza | ill give it a shot | 16:19 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for Ocata 2016-10-12 https://review.openstack.org/380325 | 16:19 |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 16:20 |
| evrardjp | the con of doing this, is we introduce a non-fully scoped dependency, and we need the deployers to ansible-galaxy (or clone, or whatever) the roles before starting it, where if you put a full dependency in your meta, it can be auto-fetched | 16:20 |
| evrardjp | but we do this way everywhere so it's fine :D | 16:21 |
| odyssey4me | evrardjp but only fetched if the role is actually in ansible galaxy | 16:21 |
| *** hughmFLEXin has quit IRC | 16:21 | |
| drifterza | I like that we do openstack, and other dependancies we rely on tested galaxy roles | 16:21 |
| evrardjp | yes that's what I meant | 16:22 |
| drifterza | makes my life easier | 16:22 |
| evrardjp | drifterza: it makes everyone's life easier | 16:22 |
| evrardjp | :D | 16:22 |
| drifterza | cos monasca and its 100 dependancies for backend was making me bleak | 16:22 |
| evrardjp | by hoping overhead(dependency management) < overhead(whole role maintenance) | 16:23 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Provide example configurations https://review.openstack.org/385496 | 16:23 |
| odyssey4me | spotz ^ better? | 16:23 |
| *** chris_hultin has quit IRC | 16:24 | |
| spotz | odyssey4me: Works for me:) | 16:25 |
| odyssey4me | orsm, thanks spotz :) | 16:25 |
| spotz | :) | 16:25 |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Refactoring Monasca role to align more with OSA. https://review.openstack.org/375655 | 16:26 |
| *** fops has joined #openstack-ansible | 16:27 | |
| *** chris_hultin has joined #openstack-ansible | 16:27 | |
| odyssey4me | evrardjp can you do a quick review of https://review.openstack.org/385511 please so that I can move on to better things :p | 16:28 |
| evrardjp | it's cool when I have reviews queues but I can't finish one of them because a more urgent review takes over :p | 16:29 |
| evrardjp | ok doing it right now :p | 16:29 |
| odyssey4me | haha - you're welcome to simply queue it... but it is a really simple one | 16:30 |
| evrardjp | done | 16:31 |
| evrardjp | I'm not surprised by this bug | 16:31 |
| odyssey4me | it really makes me want to table flip | 16:31 |
| odyssey4me | but I think I shall go and buy some scotch instead :) | 16:32 |
| evrardjp | does that makes me jade, a soothsayer or just predictable? | 16:32 |
| evrardjp | odyssey4me: well that's a proper way to fix the world | 16:32 |
| *** drifterza has quit IRC | 16:35 | |
| ioni | i like the proposed documentation with example configurations in production | 16:37 |
| mhayden | odyssey4me: yessir | 16:37 |
| ioni | now i need to see if i understood the initial documentation :) | 16:37 |
| mhayden | odyssey4me: i didn't know you were fighting the problem, too | 16:38 |
| odyssey4me | ioni glad you like it :) it'd be ideal if you could validate that it's right... it was all done in theory for me - I haven't tested it | 16:38 |
| odyssey4me | mhayden I'm just glad that it's solved | 16:38 |
| ioni | odyssey4me, i for one did instead of vlan, vxlan, because i do not have access to the switch management | 16:38 |
| ioni | the network separation | 16:39 |
| mhayden | odyssey4me: yeah, i was really banging my head against a desk in the security role | 16:39 |
| odyssey4me | ioni well, it is entirely plausible that the network setup can be done without vlans even | 16:39 |
| ioni | odyssey4me, indeed | 16:39 |
| ioni | odyssey4me, but i did want to separate the networks | 16:39 |
| odyssey4me | ioni perhaps you could push up another network configuration example for a test environment *without* VLAN's ? | 16:40 |
| *** haad1 has quit IRC | 16:41 | |
| mhayden | 6355555555555555555555555555555555555555 | 16:42 |
| odyssey4me | cores - we could do with some eyes on https://review.openstack.org/#/q/starredby:jesse-pretorius+project:%255Eopenstack/openstack-ansible.*+status:open please | 16:42 |
| mhayden | oops | 16:42 |
| odyssey4me | mhayden cat walk on your keyboard? | 16:42 |
| odyssey4me | or was that you bashing your head again? | 16:43 |
| mhayden | hah, laid something across the laptop | 16:43 |
| ioni | odyssey4me, i don't have without vlan or without vlanx | 16:46 |
| ioni | i'll invent something if you want | 16:46 |
| odyssey4me | going afk for a bit | 16:47 |
| *** haad1 has joined #openstack-ansible | 16:50 | |
| *** retreved has quit IRC | 16:51 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests: Move ANSIBLE_CLI_PARAMETERS after playbook path https://review.openstack.org/385511 | 16:54 |
| *** appprod0 has joined #openstack-ansible | 16:54 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Security: Add tasks for RHEL-07-010010 https://review.openstack.org/383943 | 17:02 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Security: Remove quotes from extra vars https://review.openstack.org/385563 | 17:02 |
| mhayden | odyssey4me: now that the parameter move has merged, this *should* get the rhel7 security gate back in line -> https://review.openstack.org/#/c/385563/ | 17:03 |
| *** javeriak has joined #openstack-ansible | 17:06 | |
| *** thorst_afk is now known as thorst | 17:14 | |
| *** michaelgugino has quit IRC | 17:14 | |
| evrardjp | I'm now headached, I'm off for today... Enough stacking for a day! :p | 17:14 |
| evrardjp | see you tomorrow | 17:15 |
| *** haad1 has quit IRC | 17:17 | |
| odyssey4me | mhayden does this mean it worked? http://logs.openstack.org/63/385563/1/check/gate-openstack-ansible-security-ansible-func_rhel7-centos-7-nv/dc050b9/console.html#_2016-10-12_17_07_14_418762 | 17:21 |
| *** NachoDuck has quit IRC | 17:22 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Security: Add tasks for RHEL-07-010010 https://review.openstack.org/383943 | 17:23 |
| *** NachoDuck has joined #openstack-ansible | 17:24 | |
| *** Jack_Iv has quit IRC | 17:26 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift: Use centralised Ansible test scripts https://review.openstack.org/382314 | 17:26 |
| *** Jack_Iv has joined #openstack-ansible | 17:27 | |
| mhayden | odyssey4me: yeah, it worked... this now passes, too https://review.openstack.org/385563 | 17:28 |
| *** michaelgugino has joined #openstack-ansible | 17:28 | |
| mhayden | is 'stdout_lines' getting deprecated? | 17:29 |
| odyssey4me | mhayden not as far as I know | 17:30 |
| odyssey4me | why? | 17:30 |
| mhayden | ah, it's the 'when' getting parsed after the 'with_items' | 17:30 |
| mhayden | i just need a |default([]) in there | 17:30 |
| *** hughmFLEXin has joined #openstack-ansible | 17:32 | |
| odyssey4me | stevelle I may be wrong, but it looks like the the api paste changes are causing a kerfuffel: https://review.openstack.org/385543 | 17:32 |
| *** chhavi has quit IRC | 17:32 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible: Configure Calico specific BIRD settings in OSA https://review.openstack.org/385573 | 17:33 |
| stevelle | odyssey4me: haven't noticed that paste change and I don't know that middlware off the top of my head | 17:33 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Security: Add tasks for RHEL-07-010010 https://review.openstack.org/383943 | 17:35 |
| odyssey4me | stevelle that's a change in master after the newton release - not urgent | 17:35 |
| openstackgerrit | Merged openstack/openstack-ansible-plugins: Add get_networks command to neutron library https://review.openstack.org/385423 | 17:36 |
| stevelle | odyssey4me: as a note, we should get gnocchi bumped to 3.0 in stable/newton but I haven't had time to look to see if that was done yet | 17:36 |
| stevelle | I'll queue the ceilo change to look at it closer | 17:36 |
| *** weezS has quit IRC | 17:37 | |
| *** retreved has joined #openstack-ansible | 17:41 | |
| *** retreved has quit IRC | 17:41 | |
| *** retreved has joined #openstack-ansible | 17:42 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Provide example configurations https://review.openstack.org/385496 | 17:45 |
| odyssey4me | stevelle yep, we're already using stable/3.0 for gnocchi | 17:46 |
| odyssey4me | stevelle FYI it appears that gnocchi doesn't use the global requirements management (ie no upper constraints) - this means that any of its requirements are uncapped for every tag we release | 17:46 |
| odyssey4me | this is a bit of an issue for repeatable deployments | 17:47 |
| odyssey4me | it'd be ideal if you could try and get gnocchi to apply the global requirements process | 17:47 |
| *** evilt0ne has quit IRC | 17:49 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Provide example configurations https://review.openstack.org/385496 | 17:50 |
| stevelle | odyssey4me: I'm pretty sure there would be resistance to that. | 17:50 |
| openstackgerrit | Merged openstack/openstack-ansible-repo_build: Add cython to repo_pip_packages https://review.openstack.org/385498 | 17:50 |
| *** evilt0ne has joined #openstack-ansible | 17:50 | |
| *** evilt0ne has quit IRC | 17:50 | |
| *** evilt0ne has joined #openstack-ansible | 17:50 | |
| stevelle | they want it to be more an independent library | 17:50 |
| stevelle | odyssey4me: your concern is upper constraints? | 17:51 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone: Use centralised Ansible test scripts https://review.openstack.org/382306 | 17:53 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift: Use centralised Ansible test scripts https://review.openstack.org/382314 | 17:54 |
| *** weezS has joined #openstack-ansible | 18:00 | |
| ioni | odyssey4me, are you sure it's gnocchi 3.0? | 18:03 |
| ioni | gnocchi --version | 18:03 |
| ioni | gnocchi 2.6.0 | 18:03 |
| odyssey4me | ioni odd, https://github.com/openstack/openstack-ansible/blob/stable/newton/playbooks/defaults/repo_packages/gnocchi.yml#L36 | 18:07 |
| ioni | odyssey4me, and i did destroyed that container and the database earlier today | 18:08 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for Ocata 2016-10-12 https://review.openstack.org/380325 | 18:08 |
| ioni | lets see in their git | 18:09 |
| ioni | maybe they didn't bumped the version | 18:09 |
| *** evilt0ne has quit IRC | 18:13 | |
| *** asettle has joined #openstack-ansible | 18:14 | |
| *** seventy has joined #openstack-ansible | 18:14 | |
| *** evilt0ne has joined #openstack-ansible | 18:15 | |
| *** evilt0ne has quit IRC | 18:15 | |
| *** evilt0ne has joined #openstack-ansible | 18:15 | |
| *** hughmFLEXin has quit IRC | 18:16 | |
| ioni | odyssey4me, well, i'm an idiot :D | 18:16 |
| ioni | gnocchiclient from pip is 2.6.0 | 18:16 |
| ioni | 3.0.1 | 18:17 |
| ioni | /openstack/venvs/gnocchi-14.0.0/bin/gnocchi-metricd --version | 18:17 |
| ioni | 3.0.1 | 18:17 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [DOCS] Update manual upgrade guide https://review.openstack.org/385582 | 18:17 |
| odyssey4me | haha, you got it :) | 18:18 |
| odyssey4me | afk for a bit - need a break before I dig into some bugs | 18:18 |
| ioni | alright | 18:19 |
| *** asettle has quit IRC | 18:20 | |
| *** fguillot has joined #openstack-ansible | 18:20 | |
| *** aetaric has quit IRC | 18:23 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Security: Add tasks for RHEL-07-010020 https://review.openstack.org/385583 | 18:23 |
| *** aetaric has joined #openstack-ansible | 18:23 | |
| openstackgerrit | Merged openstack/openstack-ansible-plugins: Remove xrange for run both Python 2 and Python 3 https://review.openstack.org/385280 | 18:24 |
| openstackgerrit | Merged openstack/openstack-ansible: Set default keepalived cidr if none is provided https://review.openstack.org/385514 | 18:26 |
| openstackgerrit | Merged openstack/openstack-ansible: Update role SHAs for 14.0.0 2016-10-12 https://review.openstack.org/385370 | 18:26 |
| openstackgerrit | Merged openstack/openstack-ansible: Add documentation for known-issue with ceph client. https://review.openstack.org/385107 | 18:26 |
| *** e-vad has left #openstack-ansible | 18:26 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Execute tempest using updated method https://review.openstack.org/382211 | 18:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Execute tempest using updated method https://review.openstack.org/382211 | 18:28 |
| *** asettle has joined #openstack-ansible | 18:32 | |
| michaelgugino | someone please assign importance of https://bugs.launchpad.net/openstack-ansible/+bug/1632812 as low hanging fruit | 18:34 |
| openstack | Launchpad bug 1632812 in openstack-ansible "Remove extra_volume_types from cinder.conf" [Undecided,New] | 18:34 |
| odyssey4me | michaelgugino can you add the output of what it is now and what it should look like, with a sample config to generate it? | 18:36 |
| odyssey4me | that'll make it really easy for someone to pick up | 18:36 |
| michaelgugino | k | 18:36 |
| *** retreved has quit IRC | 18:39 | |
| openstackgerrit | Merged openstack/openstack-ansible-security: Security: Remove quotes from extra vars https://review.openstack.org/385563 | 18:40 |
| *** retreved has joined #openstack-ansible | 18:40 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift: Use centralised Ansible test scripts https://review.openstack.org/382314 | 18:40 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Security: Add tasks for RHEL-07-010260 https://review.openstack.org/385587 | 18:40 |
| mhayden | odyssey4me: thanks for the nudge on those quotes ;) | 18:41 |
| odyssey4me | mhayden np :) | 18:41 |
| odyssey4me | aint nobody got time fo' dat | 18:42 |
| mhayden | haha | 18:42 |
| odyssey4me | I'd appreciate a little review love for https://review.openstack.org/#/q/project:%255Eopenstack/openstack-ansible.*+status:open+topic:bp/central-test-repository | 18:42 |
| michaelgugino | odyssey4me: bug updated | 18:43 |
| odyssey4me | great, thanks michaelgugino | 18:44 |
| odyssey4me | I pexct that someone will pick it up fairly quickly and that evrardjp will likely mark it low-hanging-fruit | 18:44 |
| odyssey4me | typically it needs to go through the triage meeting, but I'll leave it at his discretion | 18:45 |
| odyssey4me | I'm going to dig into https://bugs.launchpad.net/openstack-ansible/+bugs?search=Search&field.assignee=jesse-pretorius this evening to see what I can figure out. | 18:45 |
| michaelgugino | Typically I would just fix it myself, but it's simple enough for a new-comer, but more complex than a spelling error, and might actually entice someone. | 18:46 |
| odyssey4me | sure, appreciate that | 18:48 |
| odyssey4me | I think we're starting to get better at doing this sort of thing. | 18:48 |
| odyssey4me | it's easy to get side-tracked into fixing little niggles like this - but if you do a bug report instead then it can be picked up by someone else, or perhaps you later | 18:49 |
| odyssey4me | that keeps the focus on what you're doing instead of being side tracked | 18:49 |
| *** asettle has quit IRC | 18:55 | |
| odyssey4me | cloudnull FYI https://bugs.launchpad.net/openstack-ansible/+bug/1632818 | 18:56 |
| openstack | Launchpad bug 1632818 in openstack-ansible "Newton: No container logs available in integrated gate jobs" [Undecided,New] | 18:56 |
| *** chris_hultin has quit IRC | 18:58 | |
| *** chris_hultin has joined #openstack-ansible | 19:00 | |
| openstackgerrit | Nish Patwa(nishpatwa_) proposed openstack/openstack-ansible-ops: Added cinder metric collection, telegraf conntrack plugin. https://review.openstack.org/379138 | 19:01 |
| *** williamcaban has quit IRC | 19:01 | |
| *** spotz is now known as spotz_zzz | 19:05 | |
| *** jmckind has joined #openstack-ansible | 19:05 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Execute tempest using updated method https://review.openstack.org/382211 | 19:19 |
| *** hughmFLEXin has joined #openstack-ansible | 19:22 | |
| *** cmart has quit IRC | 19:26 | |
| cloudnull | odyssey4me: looking into that bug now | 19:27 |
| *** hybridpollo has joined #openstack-ansible | 19:28 | |
| *** chris_hultin is now known as chris_hultin|AWA | 19:29 | |
| odyssey4me | cloudnull mgariepy I suspect that https://github.com/openstack/openstack-ansible-repo_server/commit/95ac11e15cc4daf7cdc6e31e55e0fa3df2ac35ca is causing the master builds to fail. It's causing 503's for apt package requests. | 19:30 |
| odyssey4me | you can see it consistently in the logs for https://review.openstack.org/380325 | 19:31 |
| odyssey4me | unfortunately we have no container logs, so we can't see the cache log | 19:31 |
| odyssey4me | no issues for newton, which is why I suspect that patch | 19:34 |
| cloudnull | interesting that would cause caching issues. | 19:35 |
| cloudnull | that merged 16 days ago | 19:35 |
| odyssey4me | yeah, we've still got fixed role sha's in master | 19:35 |
| cloudnull | oh. | 19:35 |
| cloudnull | that makes sense. | 19:36 |
| odyssey4me | I've been trying to move us forward for ages with all sorts of nonsense | 19:36 |
| *** AnarchyAo has joined #openstack-ansible | 19:36 | |
| *** chris_hultin|AWA is now known as chris_hultin | 19:36 | |
| cloudnull | im looking into the log bits now | 19:38 |
| cloudnull | more in a min | 19:38 |
| cloudnull | hopfully | 19:38 |
| cloudnull | i also have a node spinning up with your PR to update | 19:40 |
| cloudnull | so i'll dig into that too | 19:40 |
| *** rmelero has quit IRC | 19:40 | |
| odyssey4me | just a thought too: https://bugs.launchpad.net/openstack-ansible/+bug/1632826 | 19:40 |
| openstack | Launchpad bug 1632826 in openstack-ansible "apt/yum cache causes problems when rebuilding" [Undecided,New] | 19:40 |
| cloudnull | what else do we need to tackle for the newton release ? | 19:40 |
| *** rmelero has joined #openstack-ansible | 19:41 | |
| odyssey4me | these are the open reviews at this point: https://review.openstack.org/#/q/starredby:jesse-pretorius+project:%255Eopenstack/openstack-ansible.*+status:open | 19:41 |
| cloudnull | do we have open issues ? | 19:41 |
| odyssey4me | and the bug hit list is in the channel topc | 19:41 |
| cloudnull | ok | 19:41 |
| cloudnull | the bug list has quite a few untriaged bugs. | 19:43 |
| cloudnull | is there a hitlist of known issues that we need to resolve? | 19:43 |
| odyssey4me | cloudnull yeah, as they come in qe're itemising the greenfield related ones and dealing with them | 19:45 |
| odyssey4me | there's no shortlist really | 19:45 |
| odyssey4me | so ignore anything upgrade related, unless you think it has something to do with a fundamental issue under it | 19:46 |
| odyssey4me | feel free to pick up on trying to triage these if you like? https://bugs.launchpad.net/openstack-ansible/+bugs?search=Search&field.assignee=jesse-pretorius | 19:46 |
| odyssey4me | the first one was assigned to you | 19:46 |
| cloudnull | ok i'll try and work through some of those too | 19:48 |
| *** Jack_Iv has quit IRC | 19:49 | |
| *** johnmilton has quit IRC | 19:50 | |
| *** javeriak has quit IRC | 19:52 | |
| *** elarson has quit IRC | 19:53 | |
| *** spotz_zzz is now known as spotz | 19:54 | |
| *** appprod0 has quit IRC | 20:00 | |
| *** gouthamr has quit IRC | 20:14 | |
| *** TxGirlGeek has quit IRC | 20:16 | |
| *** Mudpuppy has quit IRC | 20:16 | |
| *** TxGirlGeek has joined #openstack-ansible | 20:16 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Execute tempest using updated method https://review.openstack.org/382211 | 20:20 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Execute tempest using updated method https://review.openstack.org/382211 | 20:21 |
| *** askb has joined #openstack-ansible | 20:22 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Execute tempest using updated method https://review.openstack.org/382211 | 20:22 |
| *** jmckind has quit IRC | 20:32 | |
| *** cmart has joined #openstack-ansible | 20:32 | |
| *** javeriak has joined #openstack-ansible | 20:32 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for Ocata 2016-10-12 https://review.openstack.org/380325 | 20:32 |
| *** haad1 has joined #openstack-ansible | 20:33 | |
| *** agireud has joined #openstack-ansible | 20:35 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for Ocata 2016-10-12 https://review.openstack.org/380325 | 20:36 |
| *** appprod0 has joined #openstack-ansible | 20:36 | |
| *** cmart has quit IRC | 20:37 | |
| odyssey4me | alright, that should work as it leaves a SHA in place for the repo_server role, but at least gets us to master for everything else | 20:37 |
| *** gouthamr has joined #openstack-ansible | 20:39 | |
| *** fxpester has quit IRC | 20:40 | |
| *** fxpester has joined #openstack-ansible | 20:40 | |
| mrda | Morning OSA | 20:41 |
| evrardjp | morning to you | 20:45 |
| evrardjp | odyssey4me: what's the issue with the repo_server? | 20:45 |
| odyssey4me | evrardjp https://bugs.launchpad.net/openstack-ansible/+bug/1632847 | 20:46 |
| openstack | Launchpad bug 1632847 in openstack-ansible "repo_server package cache causing failures" [Undecided,New] | 20:46 |
| evrardjp | well that sounds bad | 20:47 |
| odyssey4me | yeah | 20:47 |
| *** pester has joined #openstack-ansible | 20:47 | |
| evrardjp | and it's not osic | 20:49 |
| cloudnull | what? | 20:49 |
| *** seventy has quit IRC | 20:49 | |
| odyssey4me | evrardjp yes, I've confirmed by downloading the package myself and confirming with infra that the issue is not provider-side | 20:50 |
| irtermite | low blow evrardjp... low blow :( ;) | 20:50 |
| evrardjp | ok | 20:50 |
| *** fxpester has quit IRC | 20:50 | |
| odyssey4me | apparently osic has a fresh sore :p | 20:50 |
| evrardjp | not a low blow, just checking it's not the source :) | 20:50 |
| cloudnull | hahaha. | 20:50 |
| * irtermite sees that cloudnull has an "osic" pouncer too | 20:50 | |
| cloudnull | yea. blog post coming soon. | 20:50 |
| irtermite | hah | 20:51 |
| evrardjp | yes I am afraid of saying something with osic in it now | 20:51 |
| irtermite | odyssey4me: it's turning green now... it'll be ready for picture day soon | 20:51 |
| irtermite | evrardjp: pings me too | 20:51 |
| evrardjp | it's my pleasure :D | 20:52 |
| *** rmelero has quit IRC | 20:52 | |
| odyssey4me | to be fair, the osic clusters are those we get the best results from | 20:52 |
| cloudnull | odyssey4me: so i have a deployment going now, | 20:52 |
| irtermite | believe it or not... i also have a cloudnull pouncer! yup | 20:52 |
| cloudnull | and its installing gnocci | 20:52 |
| cloudnull | so i'm not seeing the package cache issue. | 20:52 |
| odyssey4me | cloudnull orly | 20:52 |
| cloudnull | TASK [os_tempest : Check if tempest flavor 202 exists] ************************* | 20:52 |
| odyssey4me | cloudnull are you deploying with the master branch for all role? | 20:52 |
| odyssey4me | cloudnull ie are you deploying with https://review.openstack.org/#/c/380325/19 | 20:53 |
| *** cmart has joined #openstack-ansible | 20:53 | |
| cloudnull | https://review.openstack.org/#/c/380325/ | 20:53 |
| cloudnull | I did a checkout of that. | 20:53 |
| odyssey4me | very specifically PS 19 | 20:53 |
| cloudnull | currently on e2c73f54a1e9b28139b5cde8272c200038e33af7 | 20:54 |
| odyssey4me | what's that SHA for? | 20:54 |
| *** weezS has quit IRC | 20:54 | |
| cloudnull | git fetch https://git.openstack.org/openstack/openstack-ansible refs/changes/25/380325/19 && git checkout FETCH_HEAD | 20:54 |
| cloudnull | that's the SHA from your checkout | 20:54 |
| odyssey4me | the repo_server role's SHA? | 20:55 |
| cloudnull | 101d613b55c1ee2ea51358812ca41e5dc06727d6 | 20:55 |
| evrardjp | cloudnull: git review -d 380325,19 too long to type? :p | 20:55 |
| *** jcrst has joined #openstack-ansible | 20:55 | |
| odyssey4me | hrmf | 20:56 |
| odyssey4me | then obviously something else is wrong | 20:56 |
| odyssey4me | without container logs we're stuck trying to figure out what | 20:56 |
| cloudnull | now my sources are the instance defaults . | 20:56 |
| cloudnull | http://nova.clouds.archive.ubuntu.com/ubuntu/ | 20:56 |
| cloudnull | so maybe its related to the mirror after all? | 20:57 |
| evrardjp | could we just disable this patch if we know it's buggy? | 20:57 |
| odyssey4me | evrardjp we don't, at this time | 20:57 |
| cloudnull | I can do another kick and lock it down to the osic mirror | 20:57 |
| odyssey4me | cloudnull well, maybe | 20:57 |
| evrardjp | well that's easy just have a patchset and recheck :p | 20:57 |
| odyssey4me | cloudnull can you confirm that the cache is configured? | 20:57 |
| odyssey4me | good plan cloudnull | 20:58 |
| cloudnull | cat /etc/apt/apt.conf.d/00apt-cacher-proxy | 20:58 |
| cloudnull | Acquire::http { Proxy "http://172.29.236.100:3142"; }; | 20:58 |
| cloudnull | its configured | 20:58 |
| cloudnull | AIO test passed -- [ Run Time = 4671 seconds || 77 minutes ] | 20:58 |
| cloudnull | so maybe something mirror + cache related. | 20:59 |
| evrardjp | odyssey4me: did you see that in a regular way? | 20:59 |
| evrardjp | 0or is that intermittent? | 20:59 |
| odyssey4me | evrardjp on that review, every time | 20:59 |
| cloudnull | and this instance is running on the same flavor and network that the OSIC gate jobs run on | 21:00 |
| cloudnull | so i dont think its net related. | 21:00 |
| *** pester has quit IRC | 21:00 | |
| *** pester has joined #openstack-ansible | 21:01 | |
| odyssey4me | cloudnull the parent for https://github.com/openstack/openstack-ansible/commit/d98eb49edc4934ac3a18d49748d2b569e9db7520 has container logs, but that patch does not | 21:02 |
| *** hughmFLEXin has quit IRC | 21:03 | |
| *** javeriak has quit IRC | 21:04 | |
| *** thorst has quit IRC | 21:05 | |
| *** woodard has quit IRC | 21:05 | |
| evrardjp | odyssey4me: Oct 12 18:28:58 localhost haproxy[16492]: Server repo_cache-back/aio1_repo_container-b00b4ae2 is DOWN, reason: Layer4 connection problem, info: "Connection refused", check duration: 0ms. 0 active and 0 backup servers left. 0 sessions active, 0 requeued, 0 remaining in queue. | 21:07 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Execute tempest using updated method https://review.openstack.org/382211 | 21:07 |
| evrardjp | on your patchset 19 | 21:08 |
| evrardjp | it doesn't sound good | 21:08 |
| cloudnull | ok found the issue with logs | 21:08 |
| cloudnull | PR inbound. | 21:08 |
| *** haad2 has joined #openstack-ansible | 21:08 | |
| evrardjp | cool you make us languish? | 21:09 |
| *** Mudpuppy has joined #openstack-ansible | 21:11 | |
| *** thorst has joined #openstack-ansible | 21:11 | |
| *** jlockwood has joined #openstack-ansible | 21:12 | |
| *** Mr_Smurf_ has joined #openstack-ansible | 21:13 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Fix container log bind mount regression https://review.openstack.org/385619 | 21:13 |
| cloudnull | odyssey4me evrardjp ^ | 21:13 |
| *** jlockwood has quit IRC | 21:14 | |
| *** retreved has quit IRC | 21:14 | |
| evrardjp | well it makes perfect sense | 21:14 |
| *** gus_ has joined #openstack-ansible | 21:14 | |
| * cloudnull off to look at the mirror issues. | 21:15 | |
| cloudnull | evrardjp: do you think that the cache is flapping ? | 21:15 |
| *** chris__hultin has joined #openstack-ansible | 21:15 | |
| evrardjp | I don't know I don't have enough facts to have a theory | 21:15 |
| evrardjp | but I see issues from haproxy side of view, which doesn't mean much because I don't have timing info | 21:16 |
| *** thorst has quit IRC | 21:16 | |
| evrardjp | it could have been attempts when the apt-mirror is not yet ready | 21:16 |
| *** odyssey4me_ has joined #openstack-ansible | 21:17 | |
| *** mgagne_ has joined #openstack-ansible | 21:17 | |
| *** aetaric_ has joined #openstack-ansible | 21:17 | |
| *** Guest7820 has joined #openstack-ansible | 21:17 | |
| evrardjp | odyssey4me: I'm definitely lagging behind bugs now :p | 21:17 |
| *** irtermit- has joined #openstack-ansible | 21:18 | |
| evrardjp | cloudnull: you mind to edit odyssey4me's comment this way we can continue with our lives? | 21:18 |
| *** odyssey4me has quit IRC | 21:18 | |
| *** odyssey4me_ is now known as odyssey4me | 21:18 | |
| evrardjp | is it "lives" or "lifes" ? | 21:18 |
| evrardjp | odyssey4me: you're flapping | 21:19 |
| odyssey4me | :p | 21:19 |
| odyssey4me | back now | 21:19 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Fix container log bind mount regression https://review.openstack.org/385619 | 21:19 |
| odyssey4me | lost connection | 21:19 |
| cloudnull | done; | 21:19 |
| *** haad1 has quit IRC | 21:19 | |
| *** chris_hultin has quit IRC | 21:19 | |
| *** aetaric has quit IRC | 21:19 | |
| *** jheroux has quit IRC | 21:19 | |
| *** palendae has quit IRC | 21:19 | |
| *** haasn has quit IRC | 21:19 | |
| *** mgagne has quit IRC | 21:19 | |
| *** LanceHaig has quit IRC | 21:19 | |
| *** goldenfri has quit IRC | 21:19 | |
| *** Mr_Smurf has quit IRC | 21:19 | |
| *** irtermite has quit IRC | 21:19 | |
| *** gus has quit IRC | 21:19 | |
| cloudnull | netsplit all the things | 21:19 |
| *** chris__hultin is now known as chris_hultin | 21:19 | |
| evrardjp | I will blame this to be the cause of our mirroring issues. | 21:19 |
| evrardjp | thanks cloudnull | 21:20 |
| *** haasn` has joined #openstack-ansible | 21:20 | |
| cloudnull | np | 21:20 |
| *** jcrst has quit IRC | 21:20 | |
| odyssey4me | if anyone has the time or inclination, here's a few patches to get reviewed: https://review.openstack.org/#/q/project:%255Eopenstack/openstack-ansible.*+status:open+topic:bp/central-test-repository | 21:20 |
| *** jcrst has joined #openstack-ansible | 21:20 | |
| odyssey4me | and a few more, related more specifically to the newton release: https://review.openstack.org/#/q/starredby:jesse-pretorius+project:%255Eopenstack/openstack-ansible.*+status:open | 21:21 |
| evrardjp | I like limechat aggregated view of channels. When there is a netsplit I see everybody leaving the openstack- channels like this one, neutron, ... I don't see anything in my log anymore with the amount of ppl netsplitting :p | 21:21 |
| evrardjp | anyway | 21:22 |
| *** xar- has quit IRC | 21:23 | |
| *** jmccrory has quit IRC | 21:23 | |
| *** adreznec has quit IRC | 21:23 | |
| *** kamtamtun has quit IRC | 21:23 | |
| *** dalees has quit IRC | 21:23 | |
| *** raddaoui has quit IRC | 21:23 | |
| *** xar- has joined #openstack-ansible | 21:23 | |
| *** javeriak has joined #openstack-ansible | 21:24 | |
| *** jimbaker has quit IRC | 21:24 | |
| *** phschwartz has quit IRC | 21:24 | |
| *** cathrichardson has quit IRC | 21:24 | |
| *** phschwartz has joined #openstack-ansible | 21:25 | |
| evrardjp | odyssey4me: about this? https://bugs.launchpad.net/openstack-ansible/+bug/1632645 I think it's time to bring idempotency into our keystone module in the future | 21:26 |
| openstack | Launchpad bug 1632645 in openstack-ansible "During upgrare M->N old endpoints are not cleaned" [Undecided,New] | 21:26 |
| evrardjp | in the meantime, ops ? | 21:26 |
| evrardjp | or upgrade procedure? | 21:26 |
| odyssey4me | not sure | 21:27 |
| odyssey4me | not thinking too hard about upgrades just yet | 21:27 |
| odyssey4me | but yes, idempotency would be great | 21:27 |
| odyssey4me | I think it's time for me to stop computering. | 21:28 |
| *** jmccrory has joined #openstack-ansible | 21:28 | |
| *** adreznec has joined #openstack-ansible | 21:28 | |
| *** kamtamtun has joined #openstack-ansible | 21:28 | |
| *** dalees has joined #openstack-ansible | 21:28 | |
| *** raddaoui has joined #openstack-ansible | 21:28 | |
| *** haasn` has quit IRC | 21:29 | |
| *** haasn has joined #openstack-ansible | 21:31 | |
| *** michaelgugino has quit IRC | 21:31 | |
| *** Jeffrey4l__ has joined #openstack-ansible | 21:34 | |
| *** Jeffrey4l_ has quit IRC | 21:35 | |
| openstackgerrit | Merged openstack/openstack-ansible: Configure Calico specific BIRD settings in OSA https://review.openstack.org/385573 | 21:38 |
| *** rmelero has joined #openstack-ansible | 21:40 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Change the common proxy cache manage tasks to be stateful https://review.openstack.org/385625 | 21:42 |
| *** haad2 has quit IRC | 21:43 | |
| odyssey4me | :) cloudnull that's pretty much what I was thinking | 21:44 |
| *** thorst has joined #openstack-ansible | 21:45 | |
| cloudnull | great minds :) | 21:45 |
| cloudnull | whats the status on making xenial voting? | 21:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Execute tempest using updated method https://review.openstack.org/382211 | 21:48 |
| odyssey4me | cloudnull well, so far xenial fails more than trusty - it times out more | 21:49 |
| odyssey4me | generally speaking we'd need to make it voting with trusty on newton, but drop trusty for ocata | 21:49 |
| *** thorst has quit IRC | 21:49 | |
| odyssey4me | but making trusty and xenial voting for the integrated gate won't work for us right now - we very rarely get both passing at once | 21:50 |
| evrardjp | seems clean and reduces the amount of bugs in our list. Good addition cloudnull | 21:50 |
| cloudnull | off to create more bugs | 21:50 |
| odyssey4me | so right now we leave the status quo - for all the roles we have both trusty and xenial voting | 21:50 |
| logan- | xenial timing out is not surprising. i was seeing consistent 1:10 - 1:15 CI builds locally on mitaka+trusty. newton+xenial is building around 1:45 - 1:50 | 21:50 |
| cloudnull | ok. | 21:50 |
| cloudnull | on my AIO tests i have only been using xenial. | 21:51 |
| evrardjp | logan-: do you have stats on n + trusty? | 21:51 |
| cloudnull | so maybe there is something making it slower in the gate. | 21:51 |
| cloudnull | 's/./?/' | 21:51 |
| odyssey4me | for newton we either need to switch xenial to voting and trusty to non voting for the integrated gate - or we need to cut the time down so we can make them both voting | 21:51 |
| logan- | evrardjp: yes... https://bugs.launchpad.net/openstack-ansible/+bug/1631690 :) | 21:52 |
| openstack | Launchpad bug 1631690 in openstack-ansible "Trusty AIO fails to start DHCP on test network" [Undecided,New] - Assigned to Jesse Pretorius (jesse-pretorius) | 21:52 |
| odyssey4me | well newton has a massive bunch more code to skip, so it's not surprising that it's slower | 21:52 |
| logan- | have not had a successful test with newton+trusty yet | 21:52 |
| evrardjp | odyssey4me: I think if have to choose I'd choose xenial for O as the only voting | 21:52 |
| *** evilt0ne has quit IRC | 21:52 | |
| odyssey4me | evrardjp for O, yes - xenial should be the only thing - we need to purge all trusty code | 21:53 |
| *** jperry has joined #openstack-ansible | 21:53 | |
| *** evilt0ne has joined #openstack-ansible | 21:53 | |
| odyssey4me | we just haven't decided yet when to hit that switch | 21:53 |
| evrardjp | that's simple: start by removing all the variables files for ubuntu 14, it's gonna break! | 21:53 |
| *** chris_hultin is now known as chris_hultin|AWA | 21:53 | |
| odyssey4me | my vote would be for after milestone 1 when the upgrade work is done | 21:54 |
| cloudnull | logan-: is that related to the agents container bug noted earlier ? | 21:54 |
| odyssey4me | cloudnull yep, both https://bugs.launchpad.net/openstack-ansible/+bugs?search=Search&field.assignee=jesse-pretorius are related | 21:54 |
| logan- | yep cloudnull it seems related to the stuff we've been seeing in the agents container | 21:54 |
| cloudnull | ok. | 21:54 |
| cloudnull | that's next on my hitlist | 21:54 |
| odyssey4me | cloudnull logan- can add your pub ssh key to some broken nodes if you like - I haven't had a chance to peruse | 21:54 |
| odyssey4me | cloudnull note that https://review.openstack.org/380325 now worked, so there's something definitely wrong in that repo_server patch | 21:55 |
| cloudnull | http://cdn.pasteraw.com/o7m0t0dtxsvj3p34uantrsq8cvturvx | 21:55 |
| *** thorst has joined #openstack-ansible | 21:55 | |
| odyssey4me | note that it still ran on osic1 | 21:55 |
| cloudnull | odyssey4me: seems odd that it would work and then not. | 21:56 |
| odyssey4me | evrardjp cloudnull mhayden can we get https://review.openstack.org/380325 merged asap so that we can get back to testing the roles as they change | 21:56 |
| odyssey4me | cloudnull odd, certainly, but at this stage it's pretty essential that we get back to testing everything properly | 21:57 |
| mhayden | odyssey4me: looking | 21:57 |
| odyssey4me | we can take a bit of time to deal with the repo_server issue - and https://review.openstack.org/385625 might just be a perfectly good thing to cover it | 21:57 |
| *** jperry has quit IRC | 21:57 | |
| odyssey4me | logan- did you see my note in https://review.openstack.org/385573 ? | 21:58 |
| logan- | yup! i do need to BP it. will get that done this evening and include the tag bump | 21:59 |
| logan- | thanks | 21:59 |
| odyssey4me | sure thing | 21:59 |
| odyssey4me | it hits none of the current code paths, so it doesn't concern me in terms of risking the release stability | 21:59 |
| *** thorst has quit IRC | 22:00 | |
| odyssey4me | two birds with one stone :p | 22:02 |
| logan- | yeah. my opinion too. i will be testing it heavily over the next months as we prep newton so if there are issues I will iterate | 22:02 |
| odyssey4me | logan- will you be executing upgrade tests? | 22:02 |
| odyssey4me | ie can we count on you to help improve the upgrade process | 22:03 |
| logan- | yes | 22:03 |
| *** elarson has joined #openstack-ansible | 22:04 | |
| logan- | except upgrade is not the top priority so it may be towards christmas until I get to that part unfortunately | 22:04 |
| odyssey4me | well, we're all volunteers here so we'll take what we can get :) | 22:05 |
| evrardjp | odyssey4me: for 325 we could use master in the ARR | 22:05 |
| evrardjp | for the repo-server | 22:05 |
| evrardjp | when the rest is merged | 22:05 |
| *** fguillot has quit IRC | 22:05 | |
| odyssey4me | evrardjp maybe - that can be a follow on patch | 22:05 |
| evrardjp | we shouldn't forget this | 22:05 |
| *** elarson has quit IRC | 22:05 | |
| evrardjp | yes sure | 22:06 |
| odyssey4me | I've been trying to get that merged for over a week now | 22:06 |
| odyssey4me | I've been to rally and back | 22:06 |
| *** elarson has joined #openstack-ansible | 22:06 | |
| odyssey4me | aint nothing gonna stop that damned patch from merging now :p | 22:06 |
| odyssey4me | we had two issues ultimately - one was we broke the repo build, which I got fixed | 22:06 |
| odyssey4me | the next was that the venvs wouldn't build because rally doesn't care about requirements management in terms of the broad openstack process | 22:07 |
| odyssey4me | to prevent that we need to solve https://bugs.launchpad.net/openstack-ansible/+bug/1631992 | 22:07 |
| openstack | Launchpad bug 1631992 in openstack-ansible "Allow venvs to be built with disparate python requirements" [High,Confirmed] | 22:07 |
| odyssey4me | but I eventually got their requirements updated, and fixed a problem with their update\ | 22:07 |
| odyssey4me | that merged this afternoon | 22:08 |
| odyssey4me | then the silly cache issues | 22:08 |
| *** johnmilton has joined #openstack-ansible | 22:08 | |
| odyssey4me | can you see me pulling my hair out yet? ;) | 22:08 |
| evrardjp | your hair? I thought it was everything else :p | 22:08 |
| evrardjp | but yes | 22:08 |
| evrardjp | we can reduce the high importance soon also on the disparate requirements :p | 22:09 |
| evrardjp | everything is gonna work out :D | 22:09 |
| odyssey4me | the disparate requirements can go down to medium now | 22:09 |
| odyssey4me | the urgency is over | 22:10 |
| odyssey4me | rally is now compliant | 22:10 |
| odyssey4me | (mostly) | 22:10 |
| *** jamesdenton has quit IRC | 22:12 | |
| *** hughmFLEXin has joined #openstack-ansible | 22:12 | |
| evrardjp | let's discuss that later when/if other stuff gets more important than this | 22:12 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Execute tempest using updated method https://review.openstack.org/382211 | 22:20 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Execute tempest using updated method https://review.openstack.org/382211 | 22:21 |
| odyssey4me | 3.14^6 patch sets later | 22:21 |
| *** jcrst has quit IRC | 22:23 | |
| *** jcrst has joined #openstack-ansible | 22:25 | |
| *** smatzek has joined #openstack-ansible | 22:27 | |
| *** TxGirlGeek has quit IRC | 22:28 | |
| *** AnarchyAo has quit IRC | 22:35 | |
| *** phalmos has quit IRC | 22:36 | |
| *** hughmFLEXin has quit IRC | 22:37 | |
| *** evilt0ne has quit IRC | 22:42 | |
| *** admin0 has quit IRC | 22:43 | |
| *** evilt0ne has joined #openstack-ansible | 22:44 | |
| openstackgerrit | Nish Patwa(nishpatwa_) proposed openstack/openstack-ansible-ops: Added cinder metric collection, telegraf conntrack plugin. https://review.openstack.org/379138 | 22:47 |
| *** smatzek has quit IRC | 22:49 | |
| *** jimbaker has joined #openstack-ansible | 22:49 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: Make ceilometer sample interval configurable https://review.openstack.org/382057 | 22:52 |
| openstackgerrit | Nish Patwa(nishpatwa_) proposed openstack/openstack-ansible-ops: Adding Kapacitor scripts [WIP] https://review.openstack.org/385174 | 22:53 |
| *** Guest7820 is now known as palendae | 22:54 | |
| openstackgerrit | Merged openstack/openstack-ansible: Update all SHAs for Ocata 2016-10-12 https://review.openstack.org/380325 | 22:57 |
| *** jimbaker has quit IRC | 23:00 | |
| cloudnull | odyssey4me: logan-: this seems like an overlayfs bug -- "tail: unrecognized file system type 0x794c7630 for ‘/var/log/neutron/neutron-ns-metadata-proxy-fd9a6190-29cc-4d75-b7e5-b1eadb36dda3.log’. please report this to bug-coreutils@gnu.org. reverting to polling" | 23:03 |
| cloudnull | I've not pin pointed it yet. | 23:03 |
| cloudnull | but working on it | 23:03 |
| odyssey4me | mhayden if you're around, a review of https://review.openstack.org/385625 would be great | 23:03 |
| odyssey4me | or evrardjp - or anyone else other than cloudnull for that matter | 23:03 |
| cjloader | odyssey4me: sorry about that, I originally did that cause I was still learning how to contribute | 23:05 |
| cjloader | odyssey4me: all good though | 23:06 |
| *** gus_ is now known as gus | 23:06 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Switch the repo_server role to master https://review.openstack.org/385684 | 23:07 |
| odyssey4me | cjloader no worries - just doing some clean-up | 23:08 |
| openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible-os_cinder: Add remote procedure call tunables to cinder https://review.openstack.org/360749 | 23:09 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Execute tempest using updated method https://review.openstack.org/382211 | 23:12 |
| cloudnull | odyssey4me logan-: interestingly the module is in the base container | 23:14 |
| *** spotz is now known as spotz_zzz | 23:14 | |
| cloudnull | ## /var/lib/lxc/ubuntu-trusty-amd64/rootfs/lib/x86_64-linux-gnu/libdbus-1.so.3 | 23:14 |
| *** jimbaker has joined #openstack-ansible | 23:15 | |
| cloudnull | which is not the case in an instance NOT using overlayfs | 23:15 |
| odyssey4me | can you load it manually? | 23:15 |
| odyssey4me | hmm, in the instance not using overlayfs does it get it from elsewhere perhaps? | 23:16 |
| *** cathrichardson has joined #openstack-ansible | 23:17 | |
| cloudnull | i thinks its just something that comes in with the net/dns utils. | 23:17 |
| cloudnull | confirming now | 23:17 |
| odyssey4me | basically I'm thinking that it tries to load it, doesn't find it, then hunts one of the paths it knows to look into and finds it elsewhere | 23:18 |
| *** BjoernT has quit IRC | 23:19 | |
| odyssey4me | cloudnull can you please put your findings in https://bugs.launchpad.net/openstack-ansible/+bug/1612412 or https://bugs.launchpad.net/openstack-ansible/+bug/1631690 (whichever is applicable) before you head to bed so that I can pick it up in the morning, or try to | 23:20 |
| openstack | Launchpad bug 1612412 in openstack-ansible "Xenial: neutron agents container unable to resolve names" [Low,Confirmed] - Assigned to Jesse Pretorius (jesse-pretorius) | 23:20 |
| openstack | Launchpad bug 1631690 in openstack-ansible "Trusty AIO fails to start DHCP on test network" [Undecided,New] - Assigned to Jesse Pretorius (jesse-pretorius) | 23:20 |
| cloudnull | its in the underlay which is then in the container on the delta | 23:21 |
| cloudnull | and i believe we see this in trusty and not xenial because we install so much shit in trusty | 23:21 |
| cloudnull | that we dont in xenial | 23:21 |
| cloudnull | basically. | 23:21 |
| cloudnull | i think the only fix for something like this is to disable overlayfs on trusty AIOs | 23:22 |
| cloudnull | or we can go find the busted package providing the lib and remove it | 23:22 |
| cloudnull | and hope that doesn't break something else. | 23:22 |
| odyssey4me | we can very easily just disable overlayfs for trusty at this point | 23:22 |
| odyssey4me | it's served its purpose | 23:22 |
| odyssey4me | we just need to restore https://review.openstack.org/351776 and make it not a test thing | 23:22 |
| odyssey4me | shall I revise that quickly? | 23:23 |
| cloudnull | let me see if i can find the package and remove just that | 23:23 |
| * cloudnull hunting | 23:23 | |
| odyssey4me | ok, otherwise feel free to restore that patch, link the two bugs and be happy | 23:23 |
| odyssey4me | well, lemme revise the patch quickly | 23:24 |
| odyssey4me | then you can opt to use it - or not | 23:24 |
| cloudnull | kk | 23:24 |
| odyssey4me | what's the best way to detect trusty instead of the kernel version? | 23:25 |
| logan- | i use ansible_distribution_release usually | 23:26 |
| odyssey4me | yeah, if that's 'trusty' - that'll do | 23:27 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts: Remove dbus packages installed in trusty defaults https://review.openstack.org/385687 | 23:29 |
| cloudnull | ^ logan- odyssey4me: makes my quick test happier. | 23:31 |
| cloudnull | trying a full stack now | 23:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Prevent overlayfs being used for Ubuntu Trusty or kernel < 3.18 https://review.openstack.org/351776 | 23:32 |
| odyssey4me | cloudnull for some reason that's showing in gerrit as 'cannot merge' | 23:33 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts: Remove dbus packages installed in trusty defaults https://review.openstack.org/385687 | 23:34 |
| cloudnull | rebased | 23:34 |
| *** rmelero has quit IRC | 23:34 | |
| cloudnull | we likely could prune that list down further | 23:34 |
| odyssey4me | cloudnull maybe good to have an upgrade release note indicating the removed packages, just in case. | 23:35 |
| cloudnull | due to all of the independent gating we should be good to go. | 23:35 |
| cloudnull | i can add that | 23:35 |
| cloudnull | ** removing packages from the base container ** I take that back, we should not prune that list. it's too far into the release to do that at this point. | 23:36 |
| cloudnull | but these packages are not used anywhere. | 23:36 |
| cloudnull | so i think it's safe. | 23:36 |
| odyssey4me | cloudnull it may be worth paring https://review.openstack.org/351776 down to only not doing overlayfs based on the kernel version again | 23:37 |
| *** rmelero has joined #openstack-ansible | 23:37 | |
| cloudnull | yea i'd key off the kernel version. | 23:38 |
| cloudnull | but in this case logan-'s install is on kernel 4.4 | 23:38 |
| cloudnull | so that wouldn't really fix the issue. | 23:38 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Prevent overlayfs being used for Ubuntu Trusty or kernel < 3.18 https://review.openstack.org/351776 | 23:38 |
| cloudnull | i think removing those packages is the key | 23:38 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Prevent overlayfs being used when kernel < 3.18 https://review.openstack.org/351776 | 23:39 |
| *** markvoelker has quit IRC | 23:39 | |
| odyssey4me | cool - so we have a double-fence here | 23:40 |
| odyssey4me | I'd rather have the surgical fix, but I'd also rather prevent an unwanted situation | 23:40 |
| *** klamath has quit IRC | 23:40 | |
| *** cathrichardson has quit IRC | 23:42 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Prevent overlayfs being used for an AIO when kernel < 3.18 https://review.openstack.org/351776 | 23:43 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Change the common proxy cache manage tasks to be stateful https://review.openstack.org/385625 | 23:43 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Switch the repo_server role to master https://review.openstack.org/385684 | 23:44 |
| *** evilt0ne has quit IRC | 23:45 | |
| *** evilt0ne has joined #openstack-ansible | 23:47 | |
| *** evilt0ne has quit IRC | 23:47 | |
| *** evilt0ne has joined #openstack-ansible | 23:47 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Prevent overlayfs being used for an AIO when kernel < 3.18 https://review.openstack.org/351776 | 23:49 |
| odyssey4me | ok, that's definitely me out for the night | 23:52 |
| odyssey4me | cheers all, chat again tomorrow | 23:52 |
| cloudnull | ttyl | 23:53 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!